The first results of the analysis of new EXOSAT observations of the DQ Her type cataclysmic variable TV Col are presented. The period of the 1–10 kev X-ray pulsation associated with the white-dwarf rotation is now established as 1911 s. The pulsations are most pronounced in the lower energy channels (1–3.5 keV). The X-ray light curve shows absorption features associated with the orbital period of the system.

As evidenced by the P Cygni profiles of their ultraviolet resonance lines, cataclysmic variables—like early-type stars—are known to have extensive, high velocity winds. Assisted by AAVSO visual data and IUE ultraviolet spectra, we present an observational and theoretical study of the P Cygni profiles of the dwarf nova HL CMa. As these profiles are dependent upon the ionization structure of the wind, we describe a model of a radiatively-driven shocked wind for cataclysmic variables, and present results for the temperature and ionization structure of the outflowing gas.

The analysis of the IUE low resolution spectra of SS Cyg in quiescence clearly shows modulations with the orbital phase of the continuum and the emission lines total area. The shape of such modulations is depending on the type of the optical outburst preceding the quiescent phase in which the UV spectra were collected. This behaviour, together with those in other wavelength regions, could be explained within the framework of the intermediate polar models.

The initial discovery of soft X-rays from Nova Muscae 1983 was followed by eight additional observations of the three brightest novae whose outburst stage coincided with the lifetime of EXOSAT satellite; namely three more observations of Nova Muscae 1983, three observations of Nova Vulpeculae 1984 # 1 (PW Vul), and two observations of Nova Vulpeculae 1984 # 2. Through these observations we sampled the soft X-ray light curve of classical novae from optical maximum to ~ 900 days after. The observations seem best explained by the constant bolometric luminosity model of a hot white dwarf remnant. Although the measurements suffer from limited statistics, very broad energy bandpass, and incomplete sampling of any single nova, their constraints on the theories of nova outburst are significant. One constraint is that the lifetime of the white dwarf remnant in Nova Muscae 1983 is ~ 2 to 3 years, which leads to the conclusion that the burned envelope mass Mburn should be of the order of . The second constraint is that the maximum temperature, of the white dwarf remnant should approximately be within 200 000 K to 400 000 K. We estimate that a white dwarf remnant evolving like the central star of a planetary nebula, with core mass of 0.8 to 0.9 M⊙, core luminosity of ~ 2 × 104L⊙, and envelope mass of 10−6M⊙, can explain the general characteristics of the X-ray measurements for Nova Muscae 1983. In order to have ≥ 1.1 M⊙ core mass, estimated from the early observations of bolometric luminosity in the UV to infrared range, a wind with Ṁ ≤ 5 × 10−7M⊙yr−1 appears to be necessary. The few observations of Nova Vulpeculae 1984 # 1 and Nova Vulpeculae 1984 # 2 , during the first year after outburst, give a risetime and intensity that is consistent with a constant bolometric luminosity model.

The 805 sec pulsing X-ray source H2252−035 has been observed for 7 h on September 14/15 and on September 17, 1983 in X-rays with the low energy telescope and the medium energy detectors of EXOSAT. While below 2 keV the semiamplitude of the 805 s pulses is ~ 100% in the 2.3–7.9 keV band it is only ~ 40%. X-ray dips that are more pronounced in low energies occur simultaneously with the orbital minimum of the optical light curve. The medium energy spectra during dips with respect to the non dip spectrum can be explained by just enhanced cold gas absorption of an additional absorbing column of 2 1022 cm−2. Model spectra for the 805 s minimum have to include a strong iron emission line at 6.55 keV with an equivalent width of 3 keV in addition to a reduced continuum intensity (radiating area) and enhanced low energy absorption.

Three aspects of mass transfer instability models of dwarf novae are examined. The hydrodynamic development of instabilities in the secondary are examined within Roche geometry and shown to extend at least a few degrees away from the line of centres. The form of the outburst light curves observed in SS Cygni are shown to be a natural consequence of mass transfer bursts with a duration either less than, or greater than, the disk viscous timescale. Finally the two-dimensional structure of the disc in the plane of the orbit is studied. As with α-disks the viscous evolution time following a burst of mass transfer determines the size of viscosity within the disk. Significant deviations from axial symmetry are found to be present.

Recent development in the theory of the boundary layer is considered. The model for the formation of a corona above the WD is discussed in some detail.

The interaction of the boundary layer with the stellar interior and possible mixing between the hydrogen-rich accreted matter and the core material is discussed in view of recent numerical simulations of the boundary layer.

The importance of the boundary layer has been known for some time, (see Lynden-Bell and Pringle. 1974). Yet this region of the disc has never been studied in great depth. We present here some calculations which are undertaken in order to explore some of the complex processes which can go on in this region. It is shown how the structure of the boundary layer is affected by viscosity, how oscillations can occur in the outer disk and boundary layer regions. We also show how they disperse and dissipate.

Nearly half of the total available accretion energy can be released in the boundary layer (BL) if the accreting object is slowly rotating. The spectral distribution of the emitted radiation depends crucially on the internal structure of the BL. Up to now no detailed models concerning the BL exist. We have developed an explicit two-dimensional numerical method written for axisymmetric accretion flows including viscosity effects and angular momentum. We display our first models concerning the BL structure incorporating variation of the stellar rotation and of the fraction of the released energy. The first results show a strong dependence of the BL structure on the local rate of cooling and on the rotation of the primary.

We present results of new vertical structure computations for cool, convective accretion disks. The influence of the uncertainty in the low temperature opacities on the cool branch of the viscosity-surface density relation is investigated. We discuss how the shape of this relation allows a prediction of temperatures in the accretion disk during the onset of an outburst and we compare this with observations for VW Hydri by Hassall (1984).

We have executed fine mesh calculations which can almost fully resolve the transition front in the accretion disk of dwarf novae. Results show that the effects of thermal diffusion become very important only when the heating wave passes by. But it is unlikely that they cause drastic changes’ in the situation of wave propagation. The validity of the localized front approximation is examined. It is found that this approximation is relatively good for heating waves but it is marginal for cooling waves.

Spectra of accretion disks in dwarf novae and some novalike stars have been computed. Many simplifications to the general model of the nature of a cataclysmic variable are needed to make the system numerically tractable. The necessity, justification, and implications of such simplifications are discussed together with the influences of some system parameters on the disk radiation.

We present two methods of modelling accretion disks in dwarf novae. The first one determines self-consistently the structure of a disk together with the radiation field. The computed theoretical spectra agree well with the observations of the dwarf nova WX Hyi in quiescence as well as in superoutburst. The second method is a modification of Tylenda’s (1981) approach. We have used it to calculate models of disks around main-sequence accretors. The most interesting result is that no unique solution exists for a moderate mass flux around 10−6 M⊙yr−1. Such disks are thus probably unstable.

High-resolution Coude observations of non-axisymmetric line emission -from the dwarf nova SS Cygni are presented. By subtracting the constant line component, the asymmetric line emission responsible for the observed phase shift between the absorption and emission line radial velocity curves can be isolated. The extra emission is a large fraction of the total line emission and extends to large velocities (~1500 km sec−1). The phase stability of the emission demands a large-scale structure which is fixed in the frame of the binary. A magnetic origin of the excitation cannot De ruled out but is implausible. A simple explanation is that the accretion stream from the companion star is able to spill over the sags of the disk, introducing emission at non-circular velocities and most likely disturbing the upper layers of the accretion disk.

We present optical and IR observations of the dwarf nova OY Car during the May 1985 superoutburst. From them we find that the superhump has a temperature of ~8000K and an area of order half the size of the red dwarf or accretion disk. We also compare the behaviour during two simultaneous optical/IR observations. Whilst the light curves in the two pass bands are similar during one observation, in the other observation they show marked differences that may be due to a cool region in the outer disk.

We present ultraviolet and X-ray observations of the eclipsing SU UMa dwarf nova OY Car early in the decline from a superoutburst. From the UV emission line spectrum and lack of X-ray eclipse, we deduce the presence of an extended coronal region.

Two-dimensional hydrodynamic models of the collision region between the stream and the disk in a cataclysmic binary are presented. The adiabatic collision results in significant mass loss from the system and produces large turbulent regions in outer parts of the disk. Radiative cooling of the collision region leeds to a semi-continuous injection of mass into the disk in the form of dense blobs of gas. In both cases the hot regions obtained as a result of the collision are over 1/6 of the circumference of the disk long and almost 40% of its radius deep, deserving to be renamed ‘hot stripes’.

We review the recent observational and theoretical studies of the nova outburst. The observational studies have not only identified a new class of novae but theoretical simulations of this class have been found to be in excellent agreement with the observations. This new class consists of outbursts occurring on ONeMg white dwarfs in close binary systems in contrast to the other outbursts which are occurring on CO white dwarfs. We also review the effects of the β -unstable nuclei and show how their presence has a major effect on the evolution.

If the accreting white dwarf increases its mass to the Chandrasekhar mass, it will either explode as a Type I supernova or collapse to form a neutron star. In fact, there is a good agreement between the exploding white dwarf model for Type I supernovae and observations. We describe various types of evolution of accreting white dwarfs as a function of binary parameters (i.e, composition, mass, and age of the white dwarf, its companion star, and mass accretion rate), and discuss the conditions for the precursors of exploding or collapsing white dwarfs, and their relevance to cataclysmic variables. Particular attention is given to helium star cataclysmics which might be the precursors of some Type I supernovae or ultrashort period X-ray binaries. Finally we present new evolutionary calculations using the updated nuclear reaction rates for the formation of O+Ne+Mg white dwarfs, and discuss the composition structure and their relevance to the model for neon novae.

We present models for Type I supernova light curves based on the explosion of partially solid white dwarfs in close binary systems. Studies of such explosions show that they leave bound remnants of different size. Our results reproduce quite well the maximun luminosities, the expansion velocities and the shape of the light curve. As the two basic papameters that govern the light curve, the ejected mass and the mass of 56Ni produced, are variable our models reproduce the slow and fast subclasses of “classical” Type I supernovae.