Shortly after orthogonal modes of polarization were observed in pulsar radio emission (Manchester et al. 1975), a series of theoretical papers (Cocke & Pacholczyk 1976, Melrose & Stoneham 1977, Melrose 1979) were quick to point out that these abrupt, 90 degree transitions in polarization position angle could be explained by a propagation effect. More specifically, the plasma above pulsar polar caps may be birefringent. The orthogonal modes separate because they have different indices of refraction. So, as Allen & Melrose (1982) point out, the plasma is somewhat analogous to a terrestrial birefringent crystal, at least in the way it affects the ray paths of the modes. With very sensitive observations of individual pulse polarization, Stinebring et al. (1984) concluded that orthogonal modes are superposed, meaning they occur simultaneously. This observation is consistent with the birefringence hypothesis. Stinebring et al. also suggested that pulsars depolarize at high radio frequency via the superposed modes. Barnard & Arons (1986) expanded on the earlier theoretical work, and proposed that, in addition to orthogonal modes, birefringence could account for the broadening of pulsar average profiles at low radio frequency and their depolarization at high frequency. The purpose of this paper is to further investigate the suggestions of Stinebring et al. and Barnard & Arons.

This paper presents the results of the measurements of linear polarization characteristics of individual pulses at frequencies 40, 60 and 103 MHz for 10 strong pulsars. We have measured total intensity, linear polarization percentage and angle along the pulse window and obtained distribution displays for these data.

Two orthogonal polarization modes are common features in the radiation of pulsars at low frequency. The energy ratio of these two modes can change significantly between different frequencies. The linear polarisation percentage has a general tendency to increase at low frequencies, but by very different factors for different components of the integrated intensity profile. The relationship between the polarization properties of PSR 0943+10 and its mode changing behaviour was studied. It is shown that the energy ratio and the frequency of occurence of two polarization modes change simultaneously with changes in the integrated intensity profile.

The position angle (PA) behaviour of linear polarization of pulsar emission is simulated after considering the relative retardation between the core and conal components, which are believed to be generated from different emission altitudes. The PA will jump 90° at some points where the complete depolarization occurs.

We explore the detailed polarisation behaviour of pulsar 0823+26 using the technique of constructing partial “mode-separated” profiles corresponding to the primary and secondary polarisation modes. The characteristics of the two polarisation modes in this pulsar are particularly interesting, both because they are anything but orthogonal (Backer & Rankin 1980), and because the secondary mode exhibits structure seen neither in the primary mode nor in the total profile.

Using the model-angle method of polarisation-mode deconstruction (Rankin et. al. 1989; Cordes et. al. 1978), we can take the following representation of the single-vector model to define the respective primary and secondary-mode PA traverses

where χ is the PA, ϕ the longitude, the maximum PA sweep rate, and A a constant which can be computed from the total PA traverse, χc between reference longitudes ±ϕc as

and which is formally equal to sin ζ/ tan α, where α and ζ are the angles that the magnetic axis and the sight line make with the rotation axis, respectively.

A marked anti-correlation between the degree of linear polarisation dL and the intensity I, has been reported for some pulsars (Manchester, Taylor & Huguenin 1975; Xilouris et. al. 1994). A detailed investigation of the variation in the total polarisation of single pulses for some pulsars has been done, with proper statistical correction applied to the linearly polarised power. This study, then, reveals very different behaviour as well as some remarkable correlations in many pulsars, which reflect directly on the nature of pulsar radio emission.

The most important correction to be applied to linear polarisation data arises from the differences in the statistical properties of the Stokes parameters Q and U, defining the linear polarisation, and the linearly polarised power The measured value of the linearly polarised power is larger than the true value, with the difference being , where σ is the measured standard deviation in the off-pulse region in any of the Stokes parameters (Rankin, Campbell & Spangler 1975).

A sample of 62 pulsar profiles has been studied with full polarization information at high radio frequencies, aiming at studying the dependency of emission altitude at the deepest layers of pulsar magnetospheres. The emission is consistent with heights ranging between 1-2% of the light-cylinder-radius, while a slight radius-to-frequency mapping is evident.

A review of simultaneous and compiled spectra is presented, and a catalog of radio spectra for 340 pulsars reported. Pulsars with flux density measurements at three and more frequencies are included. Statistics of spectral indices and frequencies of the energy maximum and the break are presented, as well as the correlations with other pulsar parameters. The classification of 127 rich pulsar spectra has been made. Three main types of spectra are identified.

We report flux density measurements, polarimetric and timing observations of pulsars made at the highest radio frequencies to date, covering the widest frequency range from 1.4 GHz to 86 GHz. We find that the magnetic field maintains its dipolar structure throughout the active part of the magnetosphere, a region located close to the stellar surface and confined to a small slab of a few stellar radii. The change in width and shape of pulse profiles saturates at mm-wavelengths while the depolarization accelerates, leading to almost completely depolarized emission. Two pulsars seem to exhibit a turn-up in their flux density spectrum at mm-wavelengths.

During a single Arecibo observation in 1981 of the Crab pulsar, a profile at 4.7 GHz was recorded which appeared to contain additional components and an interpulse (IP) shifted to earlier phase. The experiment was continued at the VLA, taking advantage of its phased array mode to form a synthesized beam, which resolves out the bright Crab Nebula background. Observations were conducted between February 9 and May 27, 1994, at 0.33, 1.4, 4.9, and 8.4 GHz. Additional radio profiles presented here were recorded at Arecibo (0.43, 0.6, and 4.7GHz) and Effelsberg (2.7GHz) by Hankins & Fowler (unpublished).

In Figure 1 we have plotted a summary of normalized profiles from several radio frequencies and infrared. The VLA profiles are time aligned, while the rest are aligned to the main pulse (MP). A new component (labeled LFC) appears 36° ahead of the MP between 0.6 and 4.9 GHz, not coincident with the position of the precursor, and with a spectral index similar to that of the MP. The MP disappears at 8.4 GHz, probably due to spectral effects. The IP appears to undergo a transition in phase and flux, disappearing at 2.7 GHz and reappearing 10° earlier at 4.7 GHz with a radically different spectral index. Two high radio frequency components (labeled HFC1 and HFC2) appear at 4.9 and 8.4 GHz, and possibly at the noise level at 1.4 GHz. They have flatter spectral indices than the MP and IP and their centroid phase changes with respect to the MP – moving to later phase with increasing frequency. The infrared profile exhibits a “bump”, or third component near the same phase as HFC1 and HFC2.

We report on observations and frequency dependence analysis of the integrated profile of the millisecond pulsar PSR J2145-0705. Observations were performed at 102 MHz with the BSA radio telescope in Pushchino.

To analyze the frequency dependence of the integrated profile we supplemented our data with high frequency measurements by Bailes et al. (unpublished). For frequency dependence analysis we used the Gaussian fit separation method of presentation the integrated pulse profile as a superposition of Gaussian-shaped individual components (Wu, Xu and Rankin 1992; Foster, Fairhead and Backer 1991). A good fit is provided by four component structure, suggested by Bailes et al. 1994.

We verify the relationship proposed by Kijak and Gil (1996) for the pulsar radio emission altitudes

(see also Eq.3 in Gil & Krawczyk, 1996), using the pulse-profile Effelsberg raw data at 1.41 GHz. We measured profile pulse-widths at the lowest intensity level corresponding to 0.01% of the maximum intensity (Fig. 1b), using the polarlog-scale technique (Hankins and Fowler, 1986). We calculated opening angles (Fig. 1a) and emission altitudes (Fig. 1c) assuming that: i) pulsar radiation is narrow-band with radius-to-frequency mapping operating in the emission region, ii) pulsar emission is beamed tangentially to the dipolar magnetic field lines, iii) the extreme profile wings originate near or at the last open field lines.

Three pulsars have been detected with the Lexan (100 Å )filter of the EUVE Deep Survey telescope: the middle-aged isolated pulsars, PSR B0656+14 and Geminga, plus the aged millisecond pulsar PSR J0437–4715. The EUV detections have been used both alone and in combination with observations in other bandpasses to limit the physical state of the objects and to test theories regarding neutron star cooling and re-heating mechanisms. For Geminga, the EUVE data suggests the presence of an optical cyclotron spectral feature superimposed on the Rayleigh-Jeans continuum tail from a hot pulsar surface. We summarize EUV pulsar observations and results.

The HST High Speed Photometer data and stroboscopic photometry data from the Asiago 1.82m telescope (1994) show a (0.0065±0.0015) magnitude weak modulation peak in Crab’s light-curve at 60 seconds. Hubble data also suggest a phase modulation peak at the same period with the amplitude (0.07 ± 0.02)°. New stroboscopic data from December 1995 Asiago observations suggests the modulation amplitude (0.0045±0.0015) magnitude, consistent with previous results.

In the course of a programme to study optical pulsars, we have observed at different occasions PSR0540-69, the 50 msec pulsar in the LMC. Like the Crab pulsar, it has been detected at X-rays, optical and radio and its braking index (∼2) determined. In this communication we report some results obtained on this pulsar, in particular the presence of narrow features in the pulse profile and compare it with a nearly simultaneous X-ray observation.

The observations were carried out at the 3.6m telescope at La. Silla observatory, Chile. The instrument consisted on a single channel photometer coupled with a. photomultiplier (GaAs photocathode). The sampling frequency was 10 kHz and the output of the photon counting system was recorded sequentially on magnetic tape. A set of filters and diaphragms were available. A standard analysis was then performed. After corrections to the solar system barycenter, the data were analyzed and the best period determined using the test (Buccheri et al, 1983).

The linear polarization of the Crab pulsar as a function of pulse phase was observed by the High Speed Photometer on the Hubble Space Telescope in March, 1993. Observations were obtained in a bandpass centered on 2770 A using a 0.25 ms sample time, corresponding to a time resolution of 0.0075 in pulse phase. The UV polarization of the pulsar [Fig. 1] is strikingly similar to that observed in the visible (cf. Smith et al. 1988). The same values of polarization and the same swing of position angle occur through the main and secondary pulses. The polarization pulse profile must be essentially wavelength independent at frequencies above the infrared.

1) PSR B1953+29, the first millisecond binary pulsar discovered, has a low mass companion with a long orbital period (see parameters in Table 1).

2) Shortly after its discovery the averaged pulse profile was observed to change shape in a discrete fashion: two different pulse profiles were alternating (Boriakoff et al 1986). Each pulse profile remained the same for typical times of days. This is reminiscent of the mode-changing behaviour of slower pulsars, however, normally such pulsars remain in one mode for only a few minutes.

3) Observations with a Cherenkov TeV Gamma-Ray detector have shown a large pulse-profile peak when the data is folded with the correct pulse period. Recent reprocessing of the original data shows the same peak (Bowden et al 1990).

4) Comparison with optical photographs revealed a bright star (20-th magnitude) located less than 1 arcsec from the pulsar positions determined with the VLA and from timing. A star of ∼O.2M⊙ is expected to be a white dwarf, at the distance of 3.5-5kpc it would be invisible.

5) Optical spectral observations of the star with the 5-meter Palomar telescope showed a reddened spectrum without any emission or absorption lines. This is typical of very hot objects.

6) CCD images obtained with the Russian 6 meter optical telescope in the Caucasus were taken to pinpoint the location of the star with accuracies better than a small fraction of an arc second. Imperfections of the CCD star images (saturation, etc.), in particular for bright 15-18 magnitude stars, had to be taken into account to pinpoint the center of the observed image. A technique was developed for finding the center of the image with an accuracy better than 0.17 arcsec, a single pixel of the CCD image. Arcs of ellipses were fitted to cross-sections of the star image at different light intensity levels. Down to very low light levels no pulsation was found.

With the results from the Compton Gamma Ray Observatory, multiwavelength pulsar studies have taken on new importance.

With the increased sensitivity of gamma-ray detectors on the Compton Gamma-Ray Observatory (CGRO) the number of presently known gamma-ray pulsars has grown. The new detections are beginning to provide clues to the origin of the high-energy radiation in the form of emerging patterns and correlations among observed quantities such as gamma-ray efficiency and spectral index vs. age. But there are still many questions about the location of the emission and its relation to the radio, optical and X-ray pulses. This paper will review models for gamma-ray emission from pulsars and will examine how well the detailed predictions of these models account for the existing observations.

The energy source of a pulsar’s non-thermal emission is the rotational energy loss of the neutron star. The rotational energy of the neutron star is transformed into the pulsar radiation by a long sequence of processes. The processes of this sequence are discussed.

Recent X-ray and γ-ray observations have detected a number of isolated spin-powered pulsars. Studies of the pulse profiles of these objects are providing a useful guide to the site of the high energy emission. The γ-ray pulses, which are closely tied to the primary population of radiating particles, seem to be an especially useful discriminant between pulsar models. With an understanding of how the pulsar luminosity and beaming factors evolve with spin parameters, pulsar population syntheses can also be used to check the predictions of high energy emission models. In turn, comparison with γ-ray sky surveys constrains the properties of the young pulsar population.