Spectra of the old nova V603 Aql taken with the HST-GHRS did not reveal resonance doublet absorption components arising in material ejected in the 1918 outburst. However absorption components, probably produced by a disk wind, were observed, though most line emission probably comes from regions associated with an accretion disk. Most strikingly, dramatic line profile variations on time-scales of not much more than 10 m were seen.

H I emission lines in the spectrum of the recurrent nova and symbiotic star RS Oph in quiescence are sometimes accompanied by blue- or red-shifted broad emission components. The displacements of the broad components from the narrow lines change from −1300 to +1200 km s−1, but possibly not continuously. The broad components seem to appear in some restricted regions on the spectra. An asymmetry between the blue- and red-shifts is noticed.

We present MERLIN maps of V1016 Cyg at 6 cm, observed on 1992 July 21 and 1995 March 23. Previously discovered bipolar structure is resolved into three distinct peaks. Evolution and proper motion of these peaks has been measured.

We have combined IUE spectra with optical spectroscopic and photometric data collected over a few orbits of the symbiotic binary CI Cyg to follow the evolution of a disk and boundary layer during a full activity cycle. Our results indicate an extended optically thin disk during quiescent periods, which evolves into an optically thick state in an eruption; a classical boundary layer at the inner edge of the disk ionizes a surrounding H II region in quiescence, and this emission fades during the rise to visual maximum in outburst. This evolution of Ṁ rivals that observed in classical CVs and pre-main sequence stars, so symbiotic systems like CI Cyg represent another opportunity to study the physics of disk accretion.

CH Cyg has begun to show symbiotic features only since 1963. Up to now four outbursts of CH Cyg have been recorded, the fifth going on at the present time. CH Cyg reveals a lot of peculiarities as compared to ‘normal’ symbiotic stars (extended inactive periods between the ‘symbiotic’ periods; the most extended active period in 1977–1986 including the ejection of bipolar radio jets; a stage of increased activity since 1992, with peculiar profiles of the Balmer emission lines etc.); see Mikolajewski, Mikolajewska & Khudyakova (1990) for further references.

Classical models for explaining the outbursts of symbiotic stars (e.g. Mikolajewska & Kenyon 1992) are unable to account for the behaviour of CH Cyg. The magnetic rotator model has been proposed and developed by Mikolajewski & Mikolajewska (1988, 1996). The model includes a wind accreting, magnetic (B ~ 107 G), rapidly rotating (P ~ 8 m) white dwarf (WD) as the hot component of the symbiotic system. A strong magnetic field prevents matter from the red giant’s wind being accreted onto the WD surface in the usual way. Instead, three stages of accretion could be considered: inactive, propeller and accretor state.

CH Cyg and MWC 560 are very peculiar symbiotic binaries consisting of an M giant and a white dwarf companion. The systems have many features in common. In particular, both show occasional eruptions with sub-Eddington luminosity, followed by the appearance of high-velocity jets. Recently, Mikołajewski, Mikołajewska & Tomov (1995) have proposed a model of activity for these systems with a rapidly rotating magnetic white dwarf accreting from the M giant’s wind. Two states of activity are possible: propeller and accretor. Although, both stars spend most time in the propeller state, occasional propeller-accretor (and vice versa) transitions may give rise to bipolar outflows and rapid jet formations. Mikołajewski et al. (1995) also proposed to distinguish a new subclass of interacting binaries – propeller stars – with CH Cyg and MWC 560 being the prototypes.

The very slow nova PU Vul was discovered by Honda and Kuwano in 1979 after the rise from B ~ 15…16 mag to a maximum B ~ 9 mag. The object has been studied by many authors (for references see Belyakina et al. 1989; Vogel & Nussbaumer 1992) but a clear picture of the system was not given. Two minima of different duration were observed in the system in the active stage, in 1980-81 and 1993-94 (Kolotilov et al. 1995). Nussbaumer & Vogel (1996) ascribed the minima to eclipses of the hot component by the cool M giant in a binary with orbital period of 4900±100 d. Our period analysis of the preoutburst B data (Liller & Liller 1979) by Stellingwerf’s (1978) method revealed a period of 4858 d. We found the third minimum in 1926-28 (lasting for 500…630 d) in these data.

The bright (mv ≈ 5) M3III star 4 Dra (HD108907, HR4765, CQ Dra) is a unique triple system combining in itself the basic structural properties of a symbiotic-like binary and of a cataclysmic one. The red giant 4 Dra A forms a wide binary (Porb ≈ 1703 d) with a hot companion 4 Dra B which seems to be a compact cataclysmic binary system (4 Dra BC, probable Porb of about 4 h).

We have monitored the 4 Dra system in UBV at the Skalnaté Pleso and Starà Lesna Observatories and at the Tartu Astrophysical Observatory during a total of 205 nights between 1989 March 12 and 1995 June 20; a few additional data points were obtained at the Bucharest Observatory.

In this paper I review the properties and behavior of low-mass X-ray binary systems (LMXBs) that contain neutron stars (NS), concentrating on the Galactic bulge sources and bursters. I describe the observed characteristics of LMXBs including their light curves, spectra, eclipses, dips, bursts, flares, pulsations, QPO, long-term periodicities and orbital period changes, and explain how fast timing results and the distinction between ‘Z’ and ‘atoll’-type sources provide the key to a unified model of LMXB behavior.

Short-term aperiodic fluctuations with 1/f-like power spectral density (PSD), where f is the frequency, are characteristics of X-ray radiation from X-ray binaries. To investigate a mechanism creating fluctuations, we construct a cellular-automaton model for accretion disks based on the concept of self organized criticality (SOC). In this model, mass accretion takes place either by an avalanche triggered when the local mass density exceeds some critical value, or by a gradual diffusion occurring regardless of the critical condition. With this model, we can reproduce the observational PSD, the distributions of peak intensities of X-ray shots, and non-random temporal distribution of the shots. Possible accretion disk models producing l/f-like X-ray fluctuations are discussed. We conclude that such disks are likely to be advection dominated.

We find evidence for the occurrence of three different types of behaviour which as their basic distinction have different X-ray brightness levels. They are found to be recurrent.

Sco X-2 is a bright Galactic bulge Z source, whose optical counterpart was discovered in 1990 [?], and observed to exhibit strong Hα emission. We have therefore obtained spectroscopic data in order to determine the orbital period and investigate the nature of the companion star. From an analysis of the Hα emission line velocities, we find a most likely orbital period of 13.94 d. This implies an evolved companion, and is consistent with the currently known orbital periods in other Z sources.

We have observed the IR counterpart of the Atoll type LMXB GX 13+1, (Naylor, Charles & Longmore 1991), from 1993 July 11…30, with the Dutch 90 cm telescope at ESO using a Gunn z (0.9–1.0 μm) filter. Over this period 110 usable images (integration time 12 m) were obtained. The data were reduced using the ESO-MIDAS/DAPHOT reduction package. A Lomb-Scargle period search was performed with the PERIOD package.

Fig. 1 shows our Gunn z band image with stars labeled as in Naylor et al. (1991). They propose their star 101 as the IR counterpart. Fig. 2 shows the light curve we obtained for this star. The star is clearly variable on a time-scale of ~ 10 d. To estimate any period more accurately we have made a Lomb-Scargle statistics periodigram (see Fig. 3); maximum power is at 12.6 d. We estimate the error to be ~ 1 day.

We have detected He I absorption from the companion star to X1822–371 and find a lower limit to its K–velocity of 230±50km s−1. We interpret the He I as arising on the X-ray heated inner face of the companion star.

Some of the most astrophysically interesting objects are found among the radio-emitting X-ray binary stars (REXRB). The class includes the well-studied objects such as SS 433, Cyg X-3 and Sco X-1. The recent discoveries of relativistic ejection of radio knots in the X-ray transients 1915+105 (Mirabel & Rodriguez 1994) and 1655–40 (Hjellming & Rupen 1995) well illustrate the extreme nature of some of these objects.

X-ray binaries are semi-detached binary stars in which matter is transfered from a more or less normal star onto a neutron star or black hole. X-ray satellites have detected large numbers of these objects (193 in a recent catalogue by van Paradijs 1995). However only a small fraction of these are known to have radio emission (e.g. Hjellming 1988).

We present MERLIN observations of the exciting X-ray transient GRS 1915+105. Previous radio observations of GRS 1915 have shown that it is a source of superluminal radio jets. The emission of anti-parallel radio knots has led to this being described as a ‘microquasar’. MERLIN observations presented here show further details of the structure. The two main features observed are a bright extension in the direction of the superluminal jet and also, an extended region of low brightness which is elongated almost at right angles to the jet. This latter feature could either be an associated nebulosity or the remains of a jet which has undergone a radical change in direction.

The Her X-1/HZ Her system consists of a neutron star with a rotation period of 1.24s orbiting a ‘normal’ stellar companion every 1.7d. Variation in optical and ultraviolet magnitude is attributed to X-ray heating on the side facing the active neutron star with some contribution from X-ray heating of the disk. The system also has a 35-day period in which the X-ray emission displays so-called main and short ‘on’ states separated by relative X-ray quiescence. The 1.7-day optical flux variations continue throughout the 35-day period. No clear consensus exists as to the origin of the 35-day period; a popular interpretation is that of occultation of the compact X-ray source by a tilted, precessing, accretion disk.

Her X-1 is an X-ray pulsar with a rotation period of 1.24 s and a binary period of 1.7d (Tananbaum et al. 1972). The 1.7 d variations in optical and ultraviolet flux are attributed to X-ray heating of the companion star and disk (e.g. Howarth & Wilson 1983, hereafter HW83). The system displays a 35 d period, attributed to the effects of a tilted, precessing, accretion disk. Optical and ultraviolet flux variations continue unchanged throughout. This work is motivated by the following reasons:

1. • The observed IUE spectra have significantly flatter slopes than those predicted by previous models (e.g. HW83).

2. • The observed strength of the Balmer jump is anomalously low compared to that expected for a normal B star (Anderson et al. 1994).

3. • HST observations obtained by Anderson et al. (1994) in order to study emission lines have yielded high quality spectra of the continuum emission from HZ Her, enabling for the first time detailed model fitting efforts.

The 35 d period in the observed X-ray and optical light curves of Her X-1 can be understood in terms of a warped accretion disk precessing in the tidal field of the companion. We propose that repulsive forces of the X-ray driven coronal wind acting on the disk causes and maintains the warp of the disk (Fig. 1). The resulting shape of the precessing disk reproduces the main features of the observed X-ray light curve [1]. Based on the calculated disk shape we model the dip structure in the X-ray observations as caused by the interaction between the accretion stream and the warped disk.

Models of Soft X-ray Transients are presented and compared with observations. The importance of inner advection-dominated flows in quiescent transient sources is discussed, as well as the problem of global stability of the standard outer accretion disc. A comparison is made with similar problems in dwarf nova models.