The most widely adopted model of pulsar radio emission is the hollow cone model, which fits much of the experimental data. The pulsar radio emission in this model is curvature radiation of relativistic particles flowing from the magnetic poles of the neutron star along a cone of open magnetic lines. The curvature radiation is amplified at the plasma frequency, therefore different radio frequencies f originate at different radii r of the emitting regions.

Exact timing measurements allow a determination of the phase shift between observations of a pulsar at different frequencies. It has become clear from these observations that a simple dipole magnetic field configuration can not explain the time lag observed for many pulsars between profiles at high frequencies (Kuz'min et al. 1986).

There are cases which might better be explained by a combination of dipole and quadrupole field components (Davies et al. 1984). We report in this paper on new pulsar time alignment observations of a number of pulsars at high and low frequencies which support the general picture outlined above.

Coordinated profile observations between 25 MHz and 5 GHz have been carried out at the Arecibo and Pushchino observatories, and aligned according to dispersion measure (DM) values determined by cross-correlating microstructure features at adjacent frequencies. The average profiles of the nearly aligned rotator PSR 0950+08 align satisfactorily using the DM value of 2.9701 pc cm–3. For PSR 1133+16 excess delay at low frequencies is observed when alignment is made using the value of DM = 4.8413 pc cm–3 determined from microstructure cross-correlations

The evolution of integrated pulse profiles from high to low frequencies does not conform exactly to the simple model of a dipole field. Several groups (Davîes et al. 1984, Kuz'min 1986, Shitov, Malofeev, and Izvekova 1988) have suggested large field twisting due to the rotation of the pulsar, which leads to a delay in pulse arrival times for low frequencies whose sources are located in the outer parts of the magnetosphere. We are, therefore, comparing profiles of about 30 pulsars, observed at frequencies from 40 to 1400 MHz. Our preliminary analysis shows that significant delay at low frequencies occurs in many pulsars and we present here some typical examples.

A question on the accuracy of determining the dispersion measure DM for PSR 0809+74 arises in connection with the discussion which has developed in recent years concerning observations of the effect of the superdispersion time delay (STD) (Shitov 1985, Shitov, Malofeev, and Izvekova 1988, Kuz'min 1985a) attributed to the twisting of the magnetic field lines or to, competing with it as to the observation manifestations, the aberration effect (Cordes 1978, Matese and Whitmire 1980, Ulyanov 1989, Ulyanov 1990). An interpretation of the available data in favor of any of these effects would be inappropriate without a quantitative analysis of the observation errors. With our aim to define the DM value and its errors more precisely, we shall analyze both the observational data available elsewhere and our results for PSR 0809+74 which, from the observational data in Shitov (1985), Shitov et al. (1986), and Kuz'min et al. (1985), has the most pronounced STD effect.

Several methods for evaluating the angle β between the magnetic moment and rotation axis of a pulsar are considered. Averaged values of this angle, estimated by different methods for 61 pulsars, decrease as the characteristic age increases. Many pulsars have small angles = 23°) and large ages ((log т) = 7.0), so the alignment of the magnetic and rotation axes must take place during the evolution of pulsars.

A new parameter K is defined which becomes very important in our study of pulsar evolution. The distribution of pulsars in the K vs. logtc diagram reveals some constraints which can best be described by theoretical curves.

The results show that the evolution of a pulsar is characterized by two different stages corresponding to the decrease of the emission-cone width and the alignment of the magnetic axis, respectively. The decay time of the alignment of the magnetic inclination is 1.5 X 107 yr. The time scale is 6 X 104 yr when the value of ρt drops to e times its final value (t → ∞). Investigation of the birth values α(0) lends support to a random orientation of the pulsar magnetic axis with respect to the spin axis. The evolution limit of ρ is shown to have a Gaussian distribution around a central value of 5.6° with σ = 2.0°. As a result of this K analysis, new methods are suggested to determine the birth values α(0), the final values of ρ0, and the pulsar's overall geometry.

We have studied the radio pulsar emission beam assuming a) a magnetic dipole field geometry and b) that the beam geometry is defined by the field lines that are not contained within the light cylinder. In general, the beam is compressed in the meridional direction. When the magnetic and rotation axes are aligned the beam is circular and as, the angle between these axes increases, the ratio of meridional to longitudinal dimension decreases monotonically to the minimum value 0.62 when the axes are orthogonal. This beam shape is thus consistent with that inferred from the study of circular polarization in average pulse profiles by Radhakrishnan and Rankin. Evidence for meridional compression is also found in the extensive observational study of Lyne and Manchester (1988).The beam evolution was determined using this data set, the beamwidth being found proportional to P–1/2, where P is the pulsar period. This relation implies that the more rapidly rotating pulsars should have larger beams, and this should aid in their detection. The more numerous, slower pulsars should have somewhat smaller beams than previously determined. This implies that the pulsar birthrate is probably close to the highest current estimates (1 in 25yr).

The evolution of the magnetic field strength plays a major role in the life history of a neutron star. In this article the observational evidence of field evolution, in particular that of field decay and magnetic alignment, are critically examined. It is concluded that the observed decay of the spindown torque on radio pulsars cannot be caused by a secular evolution of the “obliqueness” of the neutron star, as suggested by some authors. Recent observations provide a strong indication that the decay of the magnetic field strength of a neutron star may be closely related to its evolution in a binary system. Theoretical models for such an evolution are discussed.

The claims made in many statistical analyses for magnetic field decay follow from the assumption that radio luminosity declines slower than spin-down luminosity which is suggested by least-squares fits to the data. However, such fits are very sensitive to the behaviors of the fastest and slowest pulsars. If pulsar luminosities are plotted in distance-selected groups, the (radio inefficient) Crab and Vela pulsars are clearly exceptional members and the remaining pulsars are consistent with a mean fixed conversion efficiency of 10–5. Numerical simulations (and theoretical analysis) strongly suggest that the pulsar period distribution peak is caused by luminosity selection and searches limited in distance by dispersion measure, and not by some mechanism that removes old pulsars. We cannot exclude magnetic field decay with some large time constant, ≥107 years say, but only because it would make such a small difference to the expected pulsar statistics, not because it is required by the data.

It is shown that pulsar statistics are in good agreement with the theory of current (plasma) energy losses of a neutron star. Up to now two theories of pulsar braking have been considered. The first well known model is connected with the radiation of vacuum magneto-dipole waves (Ostriker and Gunn 1969). However, according to the electrodynamical theory of pulsars, the presence of a rather dense plasma in the magnetosphere of a neutron star leads inevitably to a complete suppression of magneto-dipole losses—the dynamics of the stellar slow down is determined by the longitudinal current only (Beskin, Gurevich, and Istomin 1983).

We report the results of detailed observations of the Vela pulsar glitch which occurred on 24 December 1988. The period decrease occurred without warning and took place in much less than 2 minutes. The recovery, which commenced immediately, requires a complex function to model it, including three exponential terms and a damped sinusoid. An offset between data at two observing frequencies commenced at about the glitch epoch and continued for about 40 days; this is consistent with a small increase in dispersion measure or a change in the pulsar magnetic field configuration.

Strongly magnetized neutron stars are believed to underlie a variety of astrophysical systems, although conflicting observational and theoretical evidence has led to debate on the origin and stability of these magnetic fields. Here we describe a new model of neutron star magnetic moments, assuming that the fields are generated at birth and following their evolution to ages as large as the Hubble time. With realistic thermal evolution and conductivities, isolated neutron stars will maintain large magnetic dipole fields. As suggested elsewhere field modification under mass accretion might lead to torque decay. We identify an operative mechanism for this process; the results of this unified picture are in agreement with observations of a wide range of neutron star systems.

A program of pulse arrival time observations made at Pushchino during 1978-1986 is described. The period, period derivative and position derived from the full timing solutions were obtained for 16 pulsars. Comparison of our results with others obtained during 1970-1980 by different groups has shown that the parameters for a number of pulsars have suffered a significant change.

I survey recent theoretical work on the structure of the magnetospheres of rotation-powered pulsars, within the observational constraints set by their observed spindown, their ability to power synchrotron nebulae and their ability to produce beamed collective radio emission, while putting only a small fraction of their energy into incoherent X- and gamma radiation. I find no single theory has yet given a consistent description of the magnetosphere, but I conclude that models based on a dense outflow of pairs from the polar caps, permeated by a lower density flow of heavy ions, are the most promising avenue for future research.

The shock termination of a relativistic magnetohydrodynamic wind from a pulsar is the most interesting and viable model for the excitation of the synchrotron sources observed in plerionic supernova remnants. We have studied the structure of relativistic magnetosonic shock waves in plasmas composed purely of electrons and positrons, as well as those whose composition includes heavy ions as a minority constituent by number. We find that relativistic shocks in symmetric pair plasmas create fully thermalized distributions of particles and fields downstream. Therefore, such shocks are not good candidates for the mechanism which converts rotational energy lost from a pulsar into the nonthermal synchrotron emission observed in plerions. However, when the upstream wind contains heavy ions which are a minority constituent by number density, but carry the bulk of the energy density, much of the energy of the shock goes into a downstream, nonthermal power-law distribution of positrons with energy distribution N(E)dE ∝ E–s. In a specific model presented in some detail, s = 3. These characteristics are close to those assumed for the pairs in macroscopic MHD wind models of plerion excitation. The essential mechanism is collective synchrotron emission of left-handed extraordinary modes by the ions in the shock front at high harmonics of the ion-cyclotron frequency, with the downstream positrons preferentially absorbing almost all of this radiation, mostly at their fundamental (relativistic) cyclotron frequencies. Possible applications to the models of plerions and to constraints on theories of energy loss from pulsars are briefly outlined.

Controversial interpretations of the functioning and evolution of (radio) pulsars are discussed and confronted with the constraints. It is argued that (i) polarized e± bunches are ejected from near the polar caps at high Lorentz factors γ ≳ 102, that (ii) coherent radio pulses are emitted by them during transit to the outer magnetosphere via curvature radiation, (iii) in the shape of (transverse) fan beams, that (iv) there is no indication of magnetic field decay (or buildup), and that (v) pulsars die either by alignment or by interstellar quenching.

The so-called “millisecond” pulsars (outside globular clusters) are likely to be an abundant, young population of fast-born pulsars.