The determination of masses and mass ratios of cataclysmic binaries is discussed. It is suggested that these systems started as wide binaries, which have lost substantial amounts of mass and still more substantial amounts of angular momentum.

A two-stage mechanism is proposed whereby detached dwarf systems like V471 Tau may evolve into cataclysmic binaries. The two stages are (1) angular momentum loss from the secondary due to braking by a stellar wind linked with magnetic fields, and (2) angular momentum gain by the secondary from the orbital motion, due to tidal friction. This combination may be able to remove angular momentum from short-period binaries on a timescale of ~109 yr.

Observations of δ Pic, a β Lyr type of eclipsing binary and seven dwarf novae are reported. The ultraviolet light curves of δ Pic indicate the accumulation of matter at the triangular Lagrangian points L4 and L5, the presence of a hot spot and a higher temperature for the primary star. The implication is that the cooler secondary fills its Roche lobe and is transfering mass to the primary; mass loss to the circumstellar space and possibly to the system may also be appreciable. The temperatures of dwarf novae are derived by comparing their ultraviolet spectral energy distributions with those for normal stars of luminosity classes IV and V. Piecing together observations for different objects, the SSCyg systems have temperatures of 28 500 K, 10 000 K and 9500 K, respectively, when they are at minimum, immediately before outburst and at the beginning of rise to maximum. At maximum, the temperature is 22 500 K or 17 300 K depending on the interstellar reddening correction for AR And. Immediately before outburst, there is a large excess of flux in the far ultraviolet as indicated by the large value of the ratio of flux at 1550 Å to that at 1800 Å. The observations of Z Cam during standstill gives a temperature of 14 900 K. No excess of flux in the far ultraviolet was observed during the maximum of AR And and the standstill of Z Cam.

Accretion processes and the associated emission mechanisms are discussed, with special reference to binary X-ray sources.

A numerical simulation of equilibrium gas flow in a binary system is presented for the case in which an accretion disc forms around the accreting star (primary). Preliminary results are shown for the mass ratios 0.4, 1.0 and 2.5. Two main conclusions are reached. First, the accretion disc is quite well defined and is comparable in size to the Roche lobe of the primary. Second, all but a few per cent of the mass transferred is accreted by the primary.

The dynamics of gas streams in semi-detached binaries is reviewed from the standpoint of the theory of matched asymptotic expansions. Physical concepts are emphasized, and some astronomical applications are considered. The model of impact-driven accretion disks is also discussed briefly.

A numerical simulation of mass transfer in a semi-detached binary star, modeled to resemble the dwarf nova Z Cam, indicates that the basic flow pattern rapidly (in less than an orbital period) assumes a stream-ring configuration. The effect producing the ring is radiative energy dissipation in a shock front which initially forms between gas in the stream from the Lagrangian point (L1) and gas farther upstream which, having already circuited the white dwarf, returns to collide with the L1 stream. The accumulating shocked gas settles into a circular orbit at a radius appropriate to its angular momentum. The ring is soon massive enough (and the Keplerian radius is large enough) to obstruct the stream from L1, so that the shock becomes concentrated at the intersection of stream and ring. This shock front can be identified with the hot spot observed in cataclysmic variable stars. Similar rapid ring formation can be expected in any binary in which the mass transfer is through L1 at low velocity, and the star receiving mass is small enough to avoid collision with the stream from L1.

A turbulent accretion disc generates acoustic noise which, on propagation upward into the tenuous upper layers, forms shock waves. These give rise to a corona above the disc, thereby providing the possibility to determine observationally the existence (or otherwise) of turbulence in the disc. Models for stationary coronae of this type are calculated on the basis of a simplified thermodynamic cycle in the shocks.

The manner of mass flow in close binary systems is examined with a special view to the role of the so-called critical Roche (or Jacobian) lobe taking into consideration relevant physical conditions such as radiation pressure that may affect the restricted three-body problem treatment. The mass does not necessarily flow from component one to component two through the L1 point to form a gaseous ring surrounding the latter. These considerations are applied to X-ray binaries with early-type optical components, such as Cyg X-1 (HDE 226868) and 3U 1700-37 (HD 153919). In the two bright close binary systems β Lyr and UW CMa, which are believed to be undergoing dynamic mass transfer, recent Copernicus Princeton Telescope Spectrometer observations show that the gas giving rise to the prominent ultraviolet emission lines surrounds the entire binary system rather than merely component two. Implications of these observations are also discussed.

Average mass transfer rates in 23 Algol binaries are derived by analyzing their observed period changes within the framework of the Biermann-Hall period change model. These are compared with thermal mass loss rates computed from the dimensions which the cooler (mass-losing) component had at the onset of mass outflow. The agreement is quite good and confirms a claim made earlier that the subgiant components in Algol binaries are losing mass at a rapid (thermal) rate, not at a slow (nuclear) rate.

The introduction of image tubes makes spectroscopic studies in the red and near infrared spectral regions much more efficient. Work currently done at the Lick Observatory by the Los Angeles binary star investigators is reported. Preliminary results on element abundances in the subgiant in U Sagittae indicate that there may indeed exist systematic abundance anomalies of potential relevance to the problem of the solar neutrino flux. Emission at Hα has been observed in many systems, and several types of circumstellar features are indicated. It is suggested that some Be stars and shell stars are interacting binaries. The possibility of observing very deep stellar eclipses is discussed.

Problems of relating different observational indications of the presence of circumstellar matter in binary systems are discussed, with especial reference to the observed period changes in systems whose other properties are very similar. It is suggested that some fluctuations observed in the periods of close binary systems are apparent and not to be attributed to dynamical causes. The discussion includes some account of recent activity in U Cep, and the possible causes of spectroscopic changes in binary systems are examined.

Recent observations of emission lines from circumstellar disks in Algol-type binaries have revealed new facts concerning their structure and evolution. In some cases circumstellar disks are unstable, but the nature of the instability is not known. For U Cep, the emission lines observed during the 1974 outburst can be interpreted as arising from an expanding shell of gas about the primary. Other systems which have displayed emission line variability are RW Tau, SW Cyg, W Del, AQ Peg, and VW Cyg. There is also some evidence to suggest that RY Gem has experienced an outburst similar to that of U Cep. Emission lines observed in the eclipse spectrum of RY Gem by McKellar in 1947 were greatly different in strength and profile from those observed recently and cannot be interpreted as arising from a circumstellar disk rotating about the primary.

The large light curve changes of XY UMa, observed in the last 20 yr, are interpreted as stellar starspot activity of the primary component of this binary.

From a radial velocity study of 60 spectrograms of the binary system 68u Her, systematic velocity differences have been found to exist between the He I singlet lines and the He I triplet lines of the primary component. These differences are independent of orbital phase, date of observation or spectrograph employed and can be interpreted in terms of the existence of a large fraction of 3He in the atmosphere of the primary component. A 3He/4He ratio of between 4:1 and 49:1 is suggested.

Data on the binary character of Sco X-1 are presented.

The evolutionary status of Algol systems is examined.

Observational statements about close and contact binaries are compared with the theoretical consequences of assuming that contact binaries have a common convective envelope. It is concluded that such contact systems cannot be in thermal equilibrium, and that the inefficiency of convective heat transport in the common envelope must be allowed for. Even so, current theory seems to predict about equal numbers of contact and semidetached systems of short period, in conflict with the observations.

The evolution of a system of 1.50M⊙ and 0.50M⊙ components with initial separation 3.00R⊙ is presented. The primary reaches its Roche lobe after 6.108 × 108 yr, and the mass loss rate grows quickly to ~ 1.1 × 10−7M⊙ yr−1. The deep convective envelope of the secondary contracts in response to accretion, however, and the system avoids contact, even though the main sequence mass-radius relationship crosses the Roche relationship at 0.758M⊙.

The growth of the convective envelope of the primary with mass loss prevents thermaltimescale mass transfer from reaching any stable value. The mass loss rate continues to increase as a result of the tendency of the growing convective envelope to expand with mass loss. A Bath-type pulsational instability in the mass transfer rate is manifested as a longer-period thermal instability in the hydrostatic treatment. Eventually mass loss proceeds on a dynamical timescale, reaching 1.5 × 10−4M⊙yr−1. This is followed by a detached phase with the mass ratio not yet reversed, and a second dynamical episode leaving a 0.6666M⊙ + 1.3334M⊙ binary still in a semi-detached state. The original secondary, now more massive, has a surface luminosity of 0.0538L⊙ but a nuclear luminosity of 1540L⊙, and a convective core of 1.076M⊙ separated by a radiative zone of only 0.028M⊙ from its convective envelope.

An attempt is made to follow the evolution of the secondary in the case where the stream energy, much greater than the intrinsic luminosity of the secondary, is dissipated spherically symmetrically at its surface. The outer envelope departs radically from the previous case, an optically deep radiative surface layer forming, and the convection zone splitting into two, the lower one remaining very massive. Because the mass transfer timescale is short compared with the thermal timescale of the secondary, however, it is concluded that the stream luminosity cannot significantly alter the evolution of the secondary.

Contact binary stars coupled by a common convective envelope in which the entropy is constant, the Lucy model, are unstable against mass exchange: if either component begins to transfer mass, it will continue to do so. A detailed sequence of models is calculated which follows the thermal evolution of a 2M⊙ contact binary of normal Population I abundances (X=0.70, Z = 0.02), starting at nearly equal mass. The initial instability develops into a cyclic mass-exchange with the mass fraction oscillating between 0.56≤m2/(m1 + m2)≤0.62 with a period of ~107yr. Throughout the cycle the component stars are not in thermal equilibrium. The instability is of a general nature, and such oscillating systems can satisfactorily populate the short period, red region of the period color relation for WUMa stars.