Under the name “binary and millisecond pulsars” are grouped a number of radio pulsars, forming roughly 10% of the known pulsar population, which distinguish themselves on three counts:

• — Short spin period

• — Low magnetic-field strength

• — Presence of an evolved binary companion

If one is interested in issues related to the birth and evolution of neutron stars, a promising avenue of research begins with the study of the byproducts of a Type II supernova event, namely a pulsar (PSR) and its supernova remnant (SNR). However, it has long been a matter of some embarrassment that out of the more than 600 known PSRs and the 175 catalogued SNRs there are few associations. In fact, for nearly 15 years following the discovery of PSRs only the Crab and Vela PSRs had an associated remnant. This failure has spawned something of a cottage industry to “explain away” the low incidence of PSR/SNR associations. Beaming (Frail & Moffett 1993; Narayan & Schaudt 1988), injection (Narayan 1987), field growth (Blandford, Applegate & Hernquist 1983) and black-hole formation (Chevalier 1994) have been invoked in this context at one time or another.

PSR B1259-63 is a 47-millisecond pulsar which was discovered in a high frequency survey of the galactic plane (Johnston et al. 1992a) and was subsequently found to be in a highly eccentric orbit with a main-sequence Be star known as SS 2883 (Johnston et al. 1992b). Radio observations of the pulsar led to a phase connected timing solution which predicted the epoch of periastron to be 1994 January 9 (MJD 49361.2); optical observations of the Be star led to a determination of its mass and of the size of its circumstellar disk (Johnston et al. 1994a): the star is of approximate spectral type B1e, with mass 10 M⊙ and radius 6 R⊙. If this mass is correct and the pulsar has a mass of 1.4 M⊙, then the inclination angle of the plane of the orbit with respect to the sky is 35°. This pulsar has an unusually flat radio spectrum compared to most pulsars, which makes it easily detectable up to 8.4 GHz. The narrow pulse permits dispersion and scattering measurements for studying the ionized plasma in the system. Moreover, the pulses are highly linearly polarized and permit determination of the rotation measure (RM), allowing measurements of the magnetic field along the line of sight. The 3.5-yr orbit of the pulsar around its companion thus provides us with an excellent probe of the stellar wind of the Be star over a wide frequency range.

We describe radio timing observations of the binary pulsar PSR J0045-7319 made over the past 3.3 yr. We show that a simple timing model involving a standard Keplerian orbit describes the data well; however, significant low-level systematic residuals never before seen in other binary pulsars remain. We consider various possible origins of the residuals.

Several years of timing the pulsar in the binary neutron star system M15C have yielded the masses of both stars: the total mass is MT = 2.7121(6) M⊙; the companion mass is mc = 1.36(4) M⊙; and the pulsar mass is mp = 1.35(4) M⊙. We argue that this system is not likely to have formed through accretion-induced collapse (AIC), and that the standard model also has problems in explaining the formation.

There are about a dozen extensively investigated high-mass X-ray binaries (HMXBs), including LMC X-4, Cen X-3, 4U 1700-37, SMC X-1, Cyg X-1, and Vela X-1. Bhattacharya & Van den Heuvel (1991) compiled a list of “standard” HMXBs (see table 8 of their review article) and most of them, except for Cyg X-1 and 4U 1700-37, are accreting X-ray pulsars with an early-type or a Be star companion. Cyg X-3 was long considered to be a low-mass X-ray binary (LMXB). It was, however, recently revealed from infrared observations that the companion star has characteristics of a Wolf-Rayet star and it may be a fairly massive helium star (Van Kerkwijk et al. 1992; Van Kerkwijk 1993). I shall review here some recent progress in observational studies of the “standard” HMXBs and Cyg X-3.

Recent developments in the field of low-mass X-ray binaries are briefly reviewed, with particular emphasis on a comparison between the systems that contain accreting low magnetic-field neutron stars and those that contain black-hole candidates. The possibility that inclination effects play a role in black-hole candidate phenomenology is explored.

A “giant” outburst of A 0535+262, a transient X-ray binary pulsar, was observed in 1994 February and March with the Burst and Transient Source Experiment (BATSE) onboard the Compton Gamma-Ray Observatory. During the outburst power spectra of the hard X-ray flux contained a QPO-like component with a FWHM of approximately 50% of its center frequency. Over the course of the outburst the center frequency rose smoothly from 35 mHz to 70 mHz and then fell to below 40 mHz. We compare this QPO frequency with the neutron star spin-up rate, and discuss the observed correlation in terms of the beat frequency and Keplerian frequency QPO models in conjunction with the Ghosh-Lamb accretion torque model.

ASCA, the fourth Japanese X-ray astronomy satellite, was launched by the Institute of Space and Astronautical Science (ISAS) on 1993 February 20. ASCA is designed to be a high-capability X-ray observatory (Tanaka et al. 1994). It is equipped with nested thin-foil mirrors which provide a large effective area over a wide energy range from 0.5 to 10 keV. Two different types of detectors, CCD cameras (SIS) and imaging gas scintillation proportional counters (GIS) are employed as the focal plane instruments.

Four soft X-ray transients, two with a neutron star and two with a black hole, have been detected at quiescence with ROSAT. Blackbody fits to their spectra give temperatures of 160–300 eV, and surface areas of <1 km2. The small surface area suggests that the actual spectrum may be optically thin. The companion star does not contribute significantly to the X-ray luminosity, except perhaps in the case of A 0620-00. From the observation that accretion continues at luminosity levels of ∼1033 erg s−1 it is concluded that the neutron stars in Aql X-1 and Cen X-4 have a weak magnetic field and rotate rather slowly.

The X-ray transient systems have provided the first opportunities for detailed studies of the mass losing star in low-mass X-ray binaries. During X-ray quiescence the cool star is the dominant light source in the red and near-IR. Optical spectroscopy yields the mass function (itself a lower limit to the compact-object mass), the rotational broadening leads to the mass ratio, q (assuming only that the star fills its Roche lobe), and the IR ellipsoidal light curve gives the system inclination (for high q). In such cases, a complete solution to the system parameters is possible, and this has been performed for A 0620-00 (V616 Mon) and GS 2023+338 (V404 Cyg), leading to the first accurate black-hole masses (which are in the range 10–12 M⊙).

The orbital light curves of SXTs are discussed. In principle, the orbital inclination can be determined from the ellipsoidal variations, allowing a complete solution for the component masses. Complications in this analysis are described, and the evidence for superhumps in the light curves is reviewed. Constraints on the compact-object masses derived from superhumps are described.

A bright transient X-ray source, GRO J1655-40 (X-ray Nova Scorpii 1994) was discovered with BATSE (the Burst and Transient Source Experiment) in late July 1994. More recently, the source also became a strong radio emitter, its rise in the radio being approximately anti-correlated with a decline in the hard X-ray intensity. High-resolution radio observations subsequent to this symposium showed evidence for superluminally expanding jets. Since the hard X-ray emission extends to at least 200 keV and we find no evidence of pulsations, we tentatively classify the source as a black-hole candidate. However, its hard X-ray spectrum is unusually steep (power-law photon index α ≃ −3) relative to most other black-hole candidates. In this regard, it resembles GRS 1915+105, the first galactic source to show superluminal radio jets.

Strong variable radio emission from the bright transient X-ray source GRO J1655-40 (X-ray Nova Scorpii 1994) has been detected at 843 MHz with the Molonglo Observatory Synthesis Telescope (MOST). As the hard X-ray intensity from the 1994 August outburst declined the radio output increased rapidly, reaching a peak of nearly 8 Jy some 12 days after the first X-ray peak. VLBI images obtained at this time showed two main components separating with an apparent transverse velocity > c. The evolution of the radio spectrum suggests that the time delay between X-ray and radio emissions is due, at least in part, to opacity effects associated with this expansion.

Globular clusters form ideal laboratories for studying the interactions between stellar evolution and stellar dynamics. In the past, highly exceptional systems such as X-ray binaries and later millisecond pulsars have provided us with useful diagnostic tools. However, the fate of the bulk of the more normal stars has remained less clear. At present, rapid progress is being made in our understanding of the distributions of normal stars and primordial binaries, as well as their most abundant reaction products: blue stragglers and binaries that are produced through exchange encounters with other single stars or binaries. The complexity of the network of exchange reactions is illustrated through some specific examples, such as a formation scenario for the hierarchical triple system containing the millisecond pulsar PSR B1620-26 in M4, the first triple star system ever detected in a globular cluster.

X-ray sources in globular clusters fall into two categories: the “bright” sources, with LX ∼ 1036-1038 erg s−1, and the “dim” sources, with LX ≲ 1034.5 erg s−1. The bright sources are clearly associated with accreting neutron stars in binary systems. The nature of the dim sources, however, remains in doubt. We review recent observations of globular-cluster X-ray sources with the ROSAT satellite. ROSAT detected bright sources in M31 globular clusters and greatly increased the number of dim sources known in galactic globular clusters. We discuss what these new observations have taught us about the distribution and nature of such sources, their spectral properties, and their underlying luminosity function.

AM Herculis binaries contain mass accreting magnetic white dwarfs which appear as bright X-ray sources in the ROSAT All Sky Survey. About 52 systems are presently known which allow detailed studies of the evolution of magnetic close binaries and of fundamental plasma-physical processes in the accretion region on the white dwarf.

Supersoft X-ray sources exhibit spectra that are remarkably steep, in that the ratio of low-to-high energy X rays is much larger than is characteristic of the spectra associated with the previously known classes of luminous X-ray sources. The first supersoft sources were discovered during a survey of the Large Magellanic Cloud with the EINSTEIN Observatory (Long et al. 1981). The all-sky X-ray survey carried out with ROSAT has now established that luminous supersoft X-ray sources constitute a distinct astronomical class (see, e.g., Hasinger 1994). A number of the identified optical counterparts of the supersoft X-ray sources exhibit blue continua with emission lines of H and He II (Smale et al. 1988; Pakull et al. 1988; Cowley et al. 1990), which are characteristic of accretion disks. The X-ray emission of some sources is steady, while others exhibit significant time variability. Table 1 briefly summarizes what is known thus far about the numbers and characteristics of supersoft X-ray sources (see Hasinger 1994, and references therein).

Before the X-ray surveys performed with EINSTEIN and ROSAT soft X-ray (or EUV) sources were claimed to exist (Iben 1982; Fujimoto 1982); they were looked for in the ultraviolet and indeed such sources were found in the symbiotic systems (which count as CV-like systems) and termed the hot component of symbiotics. Although the nature of these hot components has been subject to debate (in terms of either nuclear burning white dwarfs or accretion phenomena) observational facts were in many systems favouring the first scenario (Mikolajewska & Kenyon 1992). Symbiotic binaries require wide orbits in order to keep the big giant star within its Roche lobe. It is natural to look for the short-period counterpart, but it was much more difficult to detect them in the optical due to the faintness of the secondary, which is supposed to be an evolved main-sequence star or even smaller. What turns out in these systems to be predominant is the much brighter accretion disk. It was the unique chance of the satellite borne X-ray imaging instruments to discover these EUV and soft X-ray sources and with the EINSTEIN observatory the first firm candidates were found. However, complete coverage of the soft X-ray sky was needed in order to get them all and ROSAT was the instrument which mapped the whole sky.

Data obtained with ROSAT have established the “supersoft” X-ray sources (SSS) as a separate class of objects. Their unique features include temperatures of ∼30 eV and luminosities of ∼1038 erg s−1. Hasinger (1994) has recently summarized the properties of the SSS presently known in the Galaxy and the Magellanic Clouds (cf., Kahabka & Trümper 1995). The six which are thought to be LMC members are listed in Table 1 and discussed individually, based primarily on optical data we have obtained at CTIO.