A recent study has indicated that a substantial fraction of young pulsars born in supernovae have travelled outside the boundaries of their corresponding shell supernova remnants. A simple model suggests that this should not be the case, implying either that some postulated pulsar/remnant associations are false, or that pulsars are particularly difficult to detect until they have emerged from their remnants.

We compile most recent multiplicity fractions and binary separation distributions for different primary masses, including very low-mass and brown dwarf primaries, and compare them with dynamical decay models of small-N clusters. The model predictions are based on detailed numerical calculations of the internal cluster dynamics, as well as on Monte-Carlo methods. Both observations and models reflect the same trends: (1) The multiplicity fraction is an increasing function of the primary mass. (2) The mean binary separations are increasing with the system mass in the sense that very low-mass binaries have average separations around ≈ 4AU, while the binary separation distribution for solar-type primaries peaks at ≈ 40AU. M-type binary systems apparently preferentially populate intermediate separations. Similar specific energy at the time of cluster formation for all cluster masses can possibly explain this trend.

We model the L1551 IRS5 source as a system containing two protostars, each surrounded by a circumstellar disk, both encircled by a circumbinary disk, and all disks surrounded by an extended infalling flattened envelope With this composite model, we can calculate self-consistently the spectral energy distribution of the source the shape of the ice and silicate features, and the spatial intensity distributions of the envelope and disks. We compare our model results with the observations, determining the physical parameters of the disks and the envelope. We find that flattened envelope collapse models are required in order to explain the observations and that the infall rate of the envelope is much larger than the accretion rates of the binary disks.

The statistics of stellar systems of multiplicity three and higher is reviewed. They are frequent, 0.15−0.25 of all stellar systems. Some 700 multiples are expected among the 3383 stars of spectral type F, G, and K within 50 pc, while only 76 of them are actually known. Many (if not all) close binaries have distant tertiary components, indicating that angular momentum exchange within multiple systems was probably critical in forming short-period binaries. The ratio of outer to inner periods in the best-studied nearby multiples and in low-mass pre-main sequence multiples does not exceed 104 at the formation epoch; larger ratios are produced by subsequent orbital evolution. All multiples with well-defined orbits are dynamically stable, the eccentricities of outer orbits obey the empirical stability limit Pout(1 – eout)3/Pin > 5 that is more strict than current theoretical limits. Relative orientation of orbits in triple stars shows some degree of alignment, especially in weakly-hierarchical systems. The statistics support the idea that most multiple stars originated from dynamical interactions in small clusters.

Using high-resolution (~ 01), multi-epoch Very Large Array observations, we have detected orbital motions in several low-luminosity protobinary systems in the Taurus and ρ Ophiuchus molecular complexes. The masses obtained from Kepler’s third law are of the order of 0.5 to 2 M⊙, as expected for such low-mass protostars. The relatively large bolometric luminosities of these young systems corroborates the notion that protostars obtain most of their luminosity from accretion and not from nuclear reactions. In addition, in one of the sources studied (a multiple system in Taurus), a low-mass young star has shown a drastic change in its orbit after a close approach with another component of the system, presumed to be a double star. The large proper motion achieved by this low mass protostar (20 km s−1), suggests an ejection from the system.

Orbital properties of extra-solar planets are briefly recalled and compared with equivalent features of stellar binaries. Similarities and differences are discussed. Among the more than 115 extra-solar planets discovered to date, 19 are orbiting a component of a binary system. We discuss the properties of this subsample and compare them with the equivalent characteristics of planets around single stars. Differences in the mass-period-eccentricity distributions are observed: exoplanets with m2 sin i > 2 MJup and P≤ 40-100 days are in binaries and present low eccentricities. In the context of the migration scenario, these characteristics are tentatively explained in the light of recent simulations of planet-disk interactions showing an increased accretion and migration rates of planets in case an additional perturbing close stellar companion is present in the system. Finally, different observational approaches to find planets in long-period spectroscopic binaries aiming to improve the still poor available statistics are presented. An important result is the detection of a planetary companion in the HD 41004 triple system.

High-mass X-ray binaries (HMXBs) represent an important phase in the evolution of massive binary systems and provide fundamental information on the properties of the OB-star primaries and their compact secondaries (neutron star, black hole). Recent observations indicate that the neutron stars in some of these systems (Vela X-1, 4U 1700-37) are more massive than the canonical mass of 1.35 M⊙. These observations have important consequences for the equation of state at supranuclear densities and the formation mechanism(s) of neutron stars and black holes: supernovae and gamma-ray bursts. As a consequence of the supernova explosion that produced the compact star in these systems, HMXBs have high space velocities and thus are runaways. Alternatively, OB-runaway stars can be ejected from a cluster through dynamical interactions. Observations obtained with the Hipparcos satellite indicate that both scenarios are at work.

We examine the correlation of projected rotational velocities in binary systems. It is an extension of previous work (Steinitz and Pyper, 1970; Levato, 1974). An enlarged data basis and new tests enable us to conclude that there is indeed correlation between the projected rotational velocities of components of binaries. In fact we suggest that spins are already correlated.

We discuss theoretical aspects of binary and multiple star formation, focusing on their formation in stellar clusters. First, we review individual processes that may occur during the formation of binary and multiple stars: fragmentation, accretion, interactions with circumstellar discs, and dynamical interactions. We then discuss the results from a recent large-scale hydrodynamical calculation of cluster formation in which all of these processes occur simultaneously, examining the formation mechanisms and resulting properties of the binary and multiple stars in this cluster.

Globular clusters (GCs) harbour a large number of close binaries which are hard to identify optically due to high stellar densities. Observing these GCs in X-rays, in which the compact binaries are bright, diminishes the over-crowding problem. Using the new generation of X-ray observatories, it is possible to identify populations of neutron star low mass X-ray binaries, cataclysmic variables and millisecond pulsars as well as other types of binaries. We present the spectra of a variety of binaries that we have identified in four GCs observed by XMM-Newton. We show that through population studies we can begin to understand the formation of individual classes of binaries in GCs and hence start to unfold the complex evolutionary paths of these systems.