Abstract. We derive the spectral types of the cool giants in about 100 symbiotic systems. Our classification is mainly based on near IR spectra in order to avoid the contamination of the spectrum by the nebula and the hot component in the visual region. The accuracy of our spectral types is approximately one spectral subclass, similar to previous near IR classification work, and much better than visual spectral type estimates.Strong, intrinsic spectral type variations (> 2 spectral subtypes) are only seen in systems containing pulsating mira variables.We present a catalogue of spectral types for cool giants in symbiotic systems which also includes determinations taken from the literature. The catalogue gives spectral types for the cool giants in about 170 systems which is nearly the full set of confirmed symbiotics.Based on our classifications we discuss the distribution of spectral types of the cool giants in galactic symbiotic binaries. We find that the spectral types cluster strongly between M3 and M6, with a peak at M5. The distribution of systems with a mira variable component peaks even later, at spectral types M6 and M7. This is a strong bias towards late spectral types when compared to red giants in the solar neighbourhood. Also the frequency of mira variables is much larger among symbiotic giants. This predominance of very late M-giants in symbiotic systems seems to indicate that large mass loss is a key ingredient for triggering symbiotic activity on a white dwarf companion.Further we find for symbiotic systems a strong correlation between the spectral type of the cool giant and the orbital period. In particular we find a tight relation for the minimum orbital period for symbiotic systems with Based on observations obtained with the 1.52 m and 3.6 m telescopes of the European Southern Observatory (ESO), the 1.93 m telescope of the Observatoire de Haute-Provence (OHP), the 2.3 m telescope of the Australian National University (ANU) at Siding Spring, and the William Herschel Telescope (WHT) at La Palma. This research has made use of the AFOEV database, operated at CDS, France. red giants of a given spectral type. This limiting line in the spectral type -orbital period diagram seems to be equivalent with the relation R ≤ 1 /2, where R is the radius of the red giant and 1 the distance from the center of the giant to the inner Lagrangian point L 1 . This correlation possibly discloses that symbiotic stars are -with probably only one exception in our sample -well detached binary systems.

No abstract

No abstract

Abstract. We present new dynamical parameters of the AR Pav binary system. Our observations consist of a series of high resolution optical/NIR spectra from which we derive the radial velocity curve of the red giant as well as its rotation velocity. Assuming co-rotation, we determine the stellar radius (130 R ) of the red giant. Based on this we derive the red giant's luminosity and mass (2.0 M ) as well as the distance of the system (4.9 kpc). The binary mass function finally yields the companion's mass (0.75 M ) and the binary separation (1.95 AU). We find that the red giant does not fill its Roche lobe. We review the radial velocity data of Thackeray & Hutchings (1974), and compare it with our red giant's orbit. We find that their RV curves of the blue absorption system and the permitted emission lines are in anti-phase with the red giant, and that the forbidden emission lines are shifted by a quarter of a period. The blue absorptions and the permitted emission lines are associated with the hot companion but not in a straightforward way. The blue absorption system only tracks the hot component's orbital motion whilst it is in front of the red giant, whereas at other phases line blanketing by interbinary material leads to perturbations. We finally present UV light curves based on IUE archive spectra. We clearly detect eclipses in the continuum at all wavelengths. The eclipse light curves are unusual in that they show a slow and gradual decline prior to eclipse which is followed by a sharp increase after eclipse.

No abstract