Gamma-ray bursts are generally believed to originate in the vicinity of neutron stars, but the phenomenology is still not understood. In this paper we review the known characteristics of gamma bursts and give new observational results on temporal and spectral properties. We suggest that a class of repeating bursters exists that are spectrally harder than x-ray bursters but significantly softer than “classical” gamma bursts. The March 5, 1979, burst may be the prototype of this class of bursters.

The results of recent soft X-ray and optical searches for quiescent gamma ray burster counterparts are used to constrain the properties of the neutron stars responsible for bursters. Ages are restricted to the range 2×105 y and above based on temperature upper limits and theoretical cooling curves, or 107 y and above if bursters have evolved from pulsars. Velocities are greater than 20 km/s if the neutron stars are unmagnetized. Practically no main sequence star could have escaped detection in the optical/IR searches, so if the neutron stars are in binary systems, the companion is most likely a degenerate, low mass, low temperature object.

Gamma ray bursters are interpreted as nuclear explosions under the surface of neutron stars. The explosions occur after transportation of the matter with nonequilibrium composition during starquakes in outer layers where the matter becomes unstable and explodes as a result of a developing chain reaction. In the nonequilibrium matter nuclear charge Z decreases with increasing density, contrary to the equilibrium composition. Formation of a shock wave and total energy output are calculated. Parameters of the mighty burst of 5 March 1979 are estimated. Formation of observed lines in different partions of the range and a possible nature of accompanying optical bursts are discussed with the proposed model as a basis.

Recently, several binary X-ray sources have been found to be sources of ultra high energy γ-ray emission. Air shower observations indicate photon energies >∼ 1015 eV. We review the current status of observations from the source Cygnus X-3, and compare this data with that from the sources Hercules X-1, Vela X-1, and LMC X-4. Current theoretical models for the production of γ-rays and the acceleration of high energy particles are discussed and the consequences for the evolution of such systems are examined.

Thin thermal natures of a plasma temperature of several keV are observed in the diffuse X-ray spectra from the galactic ridge region by the Tenma satellite. Within the constraints imposed by the observational results, possible origins of the ridge X-ray emission are discussed. We show that unidentified young supernova remnants in a tenuous medium could be a significant contributor to the ridge emission.

1E0630+178 was the target of a 64,000 sec EXOSAT observation on March 17–18, 1985. The 1200 photons recorded were folded around the value predicted by previous Einstein '79, '81 and EXOSAT '83 data. The region between 60.20 and 60.35 sec. was searched with independent steps of 0.005 sec, computed according to the relation: step = (period)2/observations length/number of bins in the light curve. A reduced χ2 of 3.85 (9 d.o.f.) was found for 60.285 sec. The data were then divided into five segments and the same search was performed yielding for the third segment a red. χ2 of 4.33. Moreover, the a posteriori periodogram shows also the presence of harmonics at 30.14 and 15.07 (see figure) wich appear to be even more significant (red. χ2 of 5.06 and 4.37, respectively) than the effect at 60.28.

The source of the enhanced dissipation or “α” viscosity of Keplerian accretion discs is central in all putative mechanisms for large energy release by matter accreting onto condensed objects and for the basic mechanism of star formation. The circumstances of gamma burst formation on neutron stars suggest convective buoyancy as a necessary condition for the large α. This is because the total mass ≅ 1019 g necessary to supply the energy of a gamma burst derived from infall is a natural limit for the mass stored in a disc without α viscosity. This suggests that buoyancy driven convective turbulence is the source of the enhanced transport in disc evolution models (Shakura and Sunyaev 1973). In support of this conjecture we find that the maximum possible energy released by ideal friction operating on the velocity shear of a disc is twice that required to destabilize the angular momentum distribution of such a disc. The heat energy available from an α viscosity is twice that necessary to create α in the first place. Hence, a nonlinear instability–nonlinear to create convective turbulence and nonlinear to create shear viscosity heating–is sufficient to drive α. One characteristic that would prevent the formation of such an instability is degeneracy of the disc matter as it accumulates near a neutron star (Paczynski and Jaroszynski 1978). Degeneracy inhibits strong convection because a given energy release within degenerate matter results in a large temperature, and hence large energy transport without convection. Convection occurs in an accretion disc whenever the energy which is dissipated in the disc requires a superadiabatic temperature gradient for its radiative or conductive transport to the surface. Some gamma burst mechanisms require exactly such a mechanism as a degenerate disc close to the neutron star in the correct mass (≅ 1019 g), at the correct radius several times the neutron star radius, to supply gravitational energy for a gamma burst. The degenerate disc accumulates mass stably until the density is great enough that degenerate fluid viscosity evolves the disc into contact with the neutron star. The large energy released by velocity shear at contact heats the disc causing rapid evolution and a gamma burst.

We study gamma ray spectra produced by the inverse Compton upscattering of soft photons by relativistic electrons with a one dimensional momentum distribution, which is relevant to gamma ray burst if the source magnetic field is strong enough so that the synchrotron cooling time of transverse energy becomes much shorter than isotropization time via couloub or Compton collisions. We find that for high electron longitudinal temperatures the output power is strongly beamed in the momentum direction and the spectrum softens rapidly with increasing view angle from the momentum direction.

We have constructed a model for the 1979 March 5 γ-ray burst (Cline et al. 1980; Evans et al. 1980). This event is characterized by a rapid rise (≤ 250 μs), an intense flash of duration ∼0.15s, followed by a decaying, lower intensity phase which lasted a few minutes, and was modulated with a period of 8s. The timing position error box includes a supernova remnant in the LMC, which suggests a distance of 55 kpc. At this distance the energy in the event is ∼1044 erg and the peak luminosity ∼107 greater than the Eddington limit for a 1.4 M⊙ object.

We provide new distance limits for the March 5, 1979 gamma ray transient source based on the synchrotron interpretation and the lack of low frequency self-absorption cutoff. This leads to upper limits of 0.6–11 kpc for emission areas ranging from 1 squared km polar caps to a 10 km radius stellar surface.

Results of the final analysis of COS-B X-ray observations of Cygnus X-3 (more than 220 days of observations from 1975 to 1982) are presented. Variations of the 4.8h period have been investigated and a new ephemeris including previous and later (Exosat) results, is established. Both long and short-term changes in the period are apparent. The long term variation is found to be consistent with orbital changes expected from mass transfer and mass loss in a standard low-mass binary system with an accreting neutron star. Short term period fluctuations (of the order of 4 10−3 d.) are definitively present and not fully explained. A critical discussion of the recent claimed high energy gamma (E > 1012eV) detections of the source is also presented. It is argued that, if confirmed, these results are inconsistent with the typical X-ray picture of Cygnus X-3. In particular, the eclipse duration required in the Vestrand and Eichler (1982) imply a totally unphysical companion for a 4.8h orbit.

In a previous paper (Luo, 1986), we have improved the method by Katz (1982) and Fan et al (1983). We have also found some important features in a ring model with precession and nodding motion, i.e., all the higher periodical components contain the ones with n = 1,2,3. So we can call T1,T2, and T3 fundamental periods.

Reports of air showers with E > 1015 eV from Cygnus X-3, LMC X-4, Vela X-1 and Hercules X-1 have been interpreted as requiring production of neutral secondaries by cosmic rays accelerated by the compact partner in these systems. If neutral pions are the source of photons that produce the observed air showers, then charged pions must also be produced, and they will give rise to neutrinos. We consider limits that may be placed on binary systems like Cygnus X-3 in which a neutron star is a strong source of ultra-high energy (UHE) particles that produce photons, neutrinos and other secondary particles in the companion star through nuclear interactions. The highest energy neutrinos (> 1 TeV), which have the largest interaction cross sections, are absorbed deep in the companion. From a detailed numerical calculation of the hadronic cascade induced in the atmosphere of the companion star, we estimate the neutrino production spectrum from an isotropic flux of monoenergetic 1017 eV protons and we estimate the resulting neutrino absorption in the stellar core. In the case of Cyg X-3 and LMC X-4, the cosmic-ray luminosities required to produce the observed gamma rays would result in energy deposition from neutrino absorption exceeding the intrinsic stellar luminosity of the companion. Over a timescale of 104−105 yr, the star would absorb its own binding energy and be disrupted. On shorter timescales, the energy deposition will cause significant expansion of the star, perhaps leading to quenching of high-energy signals from the source. From these results, we conclude that systems requiring intense UHE proton fluxes are either very young or the companion star is not the site of observed gamma-ray production. Alternatively, if the gamma-ray source is highly variable, the proton flux requirements would be lower, providing some relaxation of the above constraints.

[See Gaisser et al. 1986, Ap. J. (Oct. 15), in press].

An apparatus design is described. It is for detecting VHE gamma ray point sources by means of the atmospheric Cerenkov technique. Obviously, the improvement of flux sensitivity and discrimination between gamma ray and isotropic proton showers is still a key problem. Of course, it is necessary to set up more observatories and to track an object continuously with several facilities. With this in mind, we decided to develop an experiment for observing VHE gamma ray sources in China. As a first step, we will set up an apparatus which consists of three 1.5 m diameter searchlight morrors at Xinglong station of Beijing Observatory, Xinglong county, Hebei province (40°. 4N, 117°.5E, altitude 940 m). The observation will start in 1988. Then, the second apparatus will be set up at Delingha station of Purple Mountain Observatory, Delingha county, Qinghai porvince (37°.22N, altitude 3204 m). Both the sites are far from air and light pollution, and have suitable meteological condition for Cerenkov light detection as well as quite convenient facilities for transportation. Some probable technical improvements are also discussed in this paper.

Cosmic ray spectrum has an intensity enhancement at energy range 1014–16 eV/nuc. Recently Takahasi et al. (1986) called an attention to chemical composition there. Although the data still contain large uncertainties, they argued an overabundance of calcium at high energies (Ca/Fe ≥ 2 above 1014 eV/nucleus) and some enhancements of medium heavy nuclei (C ∼ Ar) instead of no anomalous p, He, and Fe abundances.

A plausible approach to the theory of pulsars as cosmic ray particle accelerators is to integrate numerically the Lorentz-Dirac-equation, using the vacuum field of a rotating orthogonal magnetic dipole as a model field configuration. Typical parameter values are: angular velocity ω = 20 π/sec and magnetic dipole moment μ = 1030G cm3 (K.O. Thielheim, Proc. ESO-CERN Conf. 1986).

At the end of this symposium, I shall briefly review our current knowledge on neutron stars.