In 1983, our group published two catalogues of KM supergiants, respectively members of the LMC (Rebeirot et al., 1983) and of the SMC (Prevot et al., 1983).

The large velocity range of the HI clouds belonging to the Magellanic Stream, as found by Haynes (1979), makes it reasonable to search for some low LSR velocity neutral hydrogen possibly belonging to the Stream. This gas, if any, would be severely blended by the local galactic gas.

If the HI clouds in the Magellanic Stream are in pressure equilibrium with the hot gaseous corona of our Galaxy, the distance of each cloud can be estimated from the observed properties of a cloud: position, angular dimensions, surface density of HI and the cloud temperature. Using a dynamical model with massive corona we have found distances of about 30–40 kpc for spherical condensations in the northern part of the Magellanic Stream.

The brightest stars always receive considerable attention in observational astronomy, but why are we so interested in these most luminous, and therefore most massive stars? These stars are our first probes for exploring the stellar content of distant galaxies. Admittedly, they are only the tip of the iceberg for the whole stellar population and very interesting processes are occurring among the less massive, older stars, but the most massive stars are our first indicators for studies of stellar evolution in other galaxies. They provide the first hint that stellar evolution may have been different in a particular galaxy because they evolve so quickly. The most luminous stars also highly influence their environments via their strong stellar winds and mass loss and eventually as supernovae.

Magellanic Cloud cepheids are of special importance for studies of stellar pulsation, of stellar evolution, of the nature of the Clouds themselves and of the extragalactic distance scale. It is not possible to cover all aspects of this subject in a short review and fortunately this is probably unnecessary since a whole symposium on cepheids is planned for Toronto next year. This paper attempts the much more modest task of assessing the current status on three main topics; chemical abundances, reddenings and the P-L and P-L-C relations (including recent infrared work). Conflicting views on some of these topics have recently appeared in the literature and a survey of the situation seems rather desirable.

Long period variable stars represent the most advanced stage of asymptotic giant branch evolution prior to planetary nebula ejection or catastrophic core collapse. In the Magellanic Clouds through study of the LPVs it has been possible to identify for the first time, stars on the AGB with luminosities right up to the AGB limit (Mbol ~ −7.1) providing direct evidence that the more massive AGB stars produce supernovae. Because the stars have well defined periods, knowledge of the temperature and luminosity enables the mass to be derived. This review will highlight spectroscopic observations of the LPVs and discuss what information they provide on nucleosynthesis on the AGB in stars of different mass.

This review will focus on photometric observations of evolved stars in the Magellanic Clouds. Emphasis is placed on red and near-infrared measurements, as these allow reasonable estimates to be made of bolometric magnitude and temperature for reliable placement in the physical HR diagram. The review is divided into three parts: field stars, cluster stars, and red variables; a summary of the surveys for objects in each of these areas is also given. Particular attention is drawn to the intermediate-age stellar population, as this component appears to be the primary tracer of the star formation rate and chemical enrichment history of the Clouds.

Recent observations in both the field and the clusters of the Magellanic Clouds suggest a higher mass loss rate during or at the end of the asymptotic giant branch phase than previously supposed. Recent theoretical investigations offer an explanation for the frequency of carbon stars in the Clouds, but a rich parameter space remains to be explored, before detailed agreement can be expected.

RR Lyrae variable stars and classical novae can be used very effectively for locating and studying the old stellar populations in the Magellanic Clouds. RR Lyrae stars are found in large numbers in both Clouds at about 19m. Apart from the concentrated searches within Cloud clusters, intensive surveys for field variables have so far been made in four areas, each about 1° square. Novae have the advantage of being as bright as 10m at maximum light. They can be detected with small telescopes and surveys have covered almost the entire area of each Cloud. Needs and prospects for future surveys are discussed. Both types of object are suitable for investigating the early chemical composition of the Clouds and its subsequent enrichment with heavy elements. Studies of the old populations in the Clouds are reinforcing the view that, while some stars in the Magellanic Clouds are as old as any in our Galaxy, the major bursts of star formation came along comparatively recently.

One LMC and two SMC fields of 0.3 sq. deg. have been searched for red variables. Carbon stars of V ~ 16–17 are common and are usually of small amplitude, while the LMC alone contains numerous faint M type variables of small amplitude. M giants of small amplitude generally have much shorter periods than carbon stars. The LMC contains numerous Miras with a P-L relation similar to that of galactic Miras, while the SMC has few Miras but many bright red variables of large amplitude which have a steeper P-L relation.

The observational material for this study consists of a series of 19 I plates (IVN+RG715) of the southern half of the LMC taken by the UK Schmidt Telescope over a six year period from 1977–1983. A small region of size 28′x56′ centred on α(1950) = 5 28 50.7, δ(1950) = −69 31 56 was scanned on all plates with a PDS microdensitometer and magnitudes were derived for typically 14000 stars per scan area. All magnitudes were converted to a common system by comparison of magnitudes to a standard plate; this comparison showed the individual rms measurement error to be 0.13 mag. No photoelectric standard sequence exists in the region studied but the instrumental magnitudes have been converted to preliminary IK magnitudes by using the IK magnitudes given by Blanco, McCarthy and Blanco (1980) for some of the red stars in the field.

The pioneering work of McGonegal et al. (1982), originally employing random-phase data for the LMC near-infrared period-luminosity (P-L) relation, is refined to mean light and extended to the SMC using new observations. We use a procedure for reducing single-phase infrared observations to mean light to obtain JHK magnitudes for Magellanic Cloud Cepheids. The new <H> P-L relations are presented and discussed.

The Small Magellanic Cloud is known to contain types of short period Cepheid variable stars not yet discovered in either the Large Magellanic Cloud or, with the exception of a single star, in the Galaxy. These variables can be divided into two categories: anomalous Cepheids and Wesselink-Shuttleworth (WS) stars. The former, which have also been found in dwarf spheroidal systems and in the globular cluster NGC 5466, have periods of 0.4–3 days, but average 0.7–1.0 mag. brighter than RR Lyrae and BL Her stars of equal period. The stars we call WS stars have periods less than about 1.1 day and, at MV = −1 to −2, are brighter than anomalous Cepheids of equal period.

Accurate MK temperature and luminosity classes have been obtained for 58 Cepheids in the Galaxy and in the Magellanic Clouds from 121 spectrograms. Classification criteria for the spectral region 3700X–6000X in the spectra of F-K supergiants are described. The spectra of galactic and LMC Cepheids are found to be similar, and Cepheids in both systems appear to obey the same relation between (B-V)o and spectral type. The low E(B-V) values of Parsons (1970) and certain other recent authors are confirmed. The spectra of SMC Cepheids show slightly weaker metal lines relative to galactic Cepheid spectra, and SMC Cepheids average about 1.3 subclasses earlier in spectral type than LMC and galactic Cepheids at the same value of (B-V)o. Reddenings for both Clouds are found to be less than 0.05. Among supergiants (Mv = −4 to −8), stars of the same spectral type have the same (B-V)o regardless of luminosity class. Spectral types at minimum light are found to be later than those reported by Feast (1974) when luminosity effects are allowed for. Spectroscopic surface gravities near minimum light appear to be lower than those near maximum and slightly fainter Mv values are found at each luminosity class than those usually cited. SMC Cepheids appear to have lower surface gravities than LMC Cepheids at the same absolute magnitude, as one would expect from the lower metal abundances.

It has been found possible to obtain information on period change in data on 115 cepheid variable stars in the Magellanic Clouds (84 LMC cepheids and 31 SMC cepheids). Harvard Observatory data of the period 1910 to 1950 (collated by Payne-Gaposchkin and Gaposchkin) are combined with Dunsink Observatory observations carried out by C.J. Butler in 1966/67 and with South African Astronomical Observatory observations covering the years 1975–1977 by Martin, Thomas, Carter and Davies to derive mean periods for the intervals between the various data sets. Using these new periods in conjunction with the very accurate Harvard periods, seperate estimates of the time averaged fractional change of period per day, d/dt (ln P), with corresponding estimated errors, could be evaluated for two epochs, one around 1950 and the other around 1971. It was found that 70 stars give rates of change of period that are not significantly different from zero, that 20 stars have two values of rate of change of period that are in agreement at the two epochs (indicative of secular period change), while 22 stars give two disparate values of rate of change of period (indicative of irregular period changes).

Controversy is raging over the question which of the relations linking periods, luminosities and colours of classical cepheids should be used to the greater advantage of cosmologists in desperate need of a reliableprimary distance indicator. Adopting a cepheid evolutionary scenario of utmost simplicity, Sandage and Tammann (1969) (=ST) have introduced the concept of an universal period-luminosity-colour (PLC) relation which is expected to minimize the Malmquist bias by virtue of its small intrinsic spread; the PLC relation constitutes the only primary pillar of their controversial distance scale. More recently, Stift (1982) and Clube and Dawe (1983) have shown that there are good reasons to consider a more realistic and hence more complex picture of cepheid evolution and photometric behaviour.

In an earlier paper by Sanduleak et al. (1978) a listing was given of 102 confirmed and probable planetary nebulae in the Large Magellanic Cloud detected on objective-prism plates taken with the Curtis Schmidt telescope at Cerro Tololo. Subsequently, deeper coverage was obtained on nitrogen-baked Kodak IIIa-J plates plus GG 455 filter exposed for 90 minutes. The thin prism was again used to provide a dispersion of about 1500 Å mm−1 at Hβ and the spectra were unwidened. An additional 25 planetary nebula candidates were found on this new plate material to show the requisite characteristics, i.e. they display (a) a stellar appearance, (b) [OIII] λλ5007, 4959 strongly in emission, and (c) no evidence of a continuum.

Measures of the equivalent widths of major stellar features in the visible and UV regions of a sample of the LMC supergiants whose types range from 05 to A1 have been derived. Their low resolution UV spectra were obtained with the IUE and complemented in the visual with spectra from ESO 3.6 m and SAAO 1.95 m telescopes. The methods used are entirely numerical and as far as possible free from subjective estimates. The data for the LMC members have been compared with those obtained for their Galactic counterparts. The variation of CIV 1550 Å and Si IV 1400 Å absorptions with spectral type has been investigated. It may be that the intensity of HeI lines is smaller for the middle and late B supergiant members of the LMC as compared with those Galactic standards, otherwise there is no systematic difference within the observational errors between the Galactic and LMC supergiants in the strengths of the principal stellar features observed in the visible as a function of spectral type. Galactic 0 and early B supergiants have greater equivalent widths of Si IV λ 1400 and C IV λ 1550 absorptions than the LMC supergiants of similar spectral types, confirming earlier observations of Hutchings.

Maximum values for the inner radii at which Fe II emission line formation can take place in the winds of three Magellanic Cloud supergiants have been determined. The lie between 3 × 1013 and 3 × 1014 for R50, R82, and R126.

On 1982 Jan. 9 a brightening of 0.75 mag of the OIafpe star R127 (=HDE 269858) was detected. Subsequently extensive photometric (optical and infrared) and high resolution spectroscopic (ground-based and IUE) observations covering a period of more than one year have been carried out. These observations have shown R127 to be a new S Dor variable which is hotter than any previously detected and studied variable of this class. For its maximum phase we derived the following basic parameters: Mbol = −10.6, Teff = 16000 K, R* = 150 R⊙. The stellar wind is characterized by a small mean velocity (v = 110 km s−1), a high mass loss rate (M = 6 · 10−5 M⊙yr−1), and a decelerated velocity field. The optical spectra show strong Balmer lines with P Cygni profiles. The He I lines changed within the one year period from strong P Cygni type lines to complex line profiles with extremely wide (FWZI = 3000 km s−1) shallow emission line wings. The IUE spectra are dominated by crowded absorption lines of singly ionized metals. A very complex shell phenomenon is indicated by the multiple substructure of these ultraviolet lines. We suggest that R127 is a massive (M > 60 M⊙) Of star evolving via a short-lived S Doradus phase to a late WN star.