Next: Black Holes in Binaries Up: Other Candidates for Ejecting Previous: Cyg X-3
According to Lipunov and Nazin (1993, 1994)[118, 119], this source belongs to a new type of binary system, so-called binaries at the non-X-ray stage: such a binary system involves a massive star in a pair with a magnetized NS - ejector.
In 1977, the highly variable radio source GT 0236+610 was discovered during a survey of the galactic plane for variable radioemission (Gregory and Taylor, 1978). Based on the accurate radio position, GT 0236+610 was identified with LSI+61 303 (Gregory et al., 1979), a B0 star at a distance of 2 kpc (Frail and Hjellming, 1991) with unusually broad, double-peaked, and variable and emission lines. Based on profile analyze (Howarth, 1983), the star was classified recently as a main-sequence B4.5III ( , , , , E(B-V)=0.75 ) emission line star with a high rotational velocity, undergoing mass loss through an equatorial disk (Hutching and Crampton, 1981). The star wind velocity is estimated as km s (Howarth, 1983). The system is an X-ray source (Bignami et al., 1981) and has been suggested (Gregory and Taylor, 1978; Perotti et al., 1980) to be a radio counterpart of the COS B gamma-ray source CG 135+01 (Hermsen et al., 1977). However, QSO 0241+62 falls within the gamma observation error box and, therefore, this identification should be viewed with caution.
The radio emission from LSI+61 303 exhibits non-thermal periodic outbursts of variable amplitude with a period of days (Taylor and Gregory, 1982, 1984[188, 189]; Coe et al., 1983). Radial velocity observations are in agreement with radio period and provide support for the presence of a compact companion with 1.1- moving in a highly eccentric orbit (Gregory et al., 1979). The radio data accumulated to date also provide evidence for a strong modulation in amplitude of radio outburst with a period 1476 7 days (Gregory et al., 1989; Paredes et al., 1990). The optical observations are consistent with the radio period and have amplitude 0.1 mag (Paredes and Figueras, 1986).
In order to explain the observed parameters, a number of models have been proposed. Taylor and Gregory (1982, 1984)[188, 189] proposed a model in which periodic radio outbursts are due to variable accretion on to a compact companion in eccentric orbit ( ) with a semi-major axis of about 5 cm. Maraschi and Treves (1981) discussed a model in which the companion is a relatively young pulsar losing energy via a relativistic wind. Paredes et al. (1991) proposed an adiabatic expansion model which includes a particle injection during a certain period of time.
Lipunov and Nasin (1994) proposed a model in which the relativistic electrons injected by the pulsar are captured by the magnetosphere of the normal companion when the pulsar passes near periastron. The model was used for the data obtained by Gregory and Taylor in August 1977, August - September 1978 and August - September 1981. These data enable us to determine some parameters of the source LSI+61 303.
The model depicted is capable of explaining all observational data accumulated so far. However, it is clear that multifrequency monitoring of LSI+61 303 is necessary to check this model and to discover the proper emission of the radiopulsar.