The timing properties of the millisecond pulsar PSR J1939+2134 -- very high rotation frequency, very low time derivative of rotation frequency, no timing glitches and relatively low timing noise -- are responsible for its exceptional timing stability over decades. It has been timed by various groups since its discovery, at diverse radio frequencies, using different hardware and analysis methods. Most of this timing data is now available in the public domain in two segments, which have not been combined so far. This work analyzes the combined data by deriving uniform methods of data selection, derivation of Dispersion Measure (DM), accounting for correlation due to "red" noise, etc. The timing noise of this pulsar is very close to a sinusoid, with a period of approximately $31$ years. The main results of this work are (1) The clock of PSR J1939+2134 is stable at the level of almost one part in $10^{15}$ over about $31$ years, (2) the power law index of the spectrum of electron density fluctuations in the direction of PSR J1939+2134 is $3.86 \pm 0.04$, (3) a Moon sized planetary companion, in an orbit of semi major axis about $11$ astronomical units and eccentricity $\approx 0.2$, can explain the timing noise of PSR J1939+2134, (4) Precession under electromagnetic torque with very small values of oblateness and wobble angle can also be the explanation, but with reduced confidence, and (5) there is excess timing noise of about $8$ $\mu$s amplitude during the epochs of steepest DM gradient, of unknown cause.

Context: The Crab pulsar underwent its largest timing glitch on 2017 Nov 8. The event was discovered at radio wavelengths, and was followed at soft X-ray energies by observatories, such as XPNAV and NICER. aims: This work aims to compare the glitch behavior at the two wavelengths mentioned above. Preliminary work in this regard has been done by the X-ray satellite XPNAV. NICER with its far superior sensitivity is expected to reveal much more detailed behavior. methods: NICER has accumulated more than $301$ kilo seconds of data on the Crab pulsar, equivalent to more than $3.3$ billion soft X-ray photons. These data were first processed using the standard NICER analysis pipeline. Then the arrival times of the X-ray photons were referred to the solar system's barycenter. Then specific analysis was done to study the specific behavior outlined in the following sections, while taking dead time into account. results: The variation of the rotation frequency of the Crab pulsar and its time derivative during the glitch is almost exactly similar at the radio and X-ray energies. The following properties of the Crab pulsar remain essentially constant before and after the glitch: the total X-ray flux; the flux, widths, and peaks of the two components of its integrated profile; and the soft X-ray spectrum. There is no evidence for giant pulses at X-ray energies. However, the timing noise of the Crab pulsar shows quasi sinusoidal variation before the glitch, with increasing amplitude, which is absent after the glitch. conclusions: Even the strongest glitch in the Crab pulsar appears not to affect all but one of the properties mentioned above, at either frequency. The fact that the timing noise appears to change due to the glitch is an important clue to unravel as this is still an unexplained phenomenon.

Glitches are important to understand the internal structure of neutron stars. They are studied using timing observations. The best studied neutron star in this respect is the Crab Pulsar. The first glitch recorded in this pulsar occurred in 1969 Sep, at an epoch when timing observations were still in their infancy, the regularity of the observations was relatively poor, and errors on the observations were relatively high. Lyne et al. (1993) analyzed most of the available data using modern techniques, and showed that this was a typical glitch of the Crab pulsar, with typical glitch parameters. This work analyses all available data, and shows that the 1969 event in the Crab pulsar is amenable to radically different interpretations. The Crab pulsar was timed by five different groups during this epoch, one at radio and the rest at optical frequencies. These data are available in the public domain, and have been analyzed using the TEMPO2 software. The 1969 event in the Crab pulsar can be better modeled as a typical glitch that was interrupted by a non-glitch speed-up event. This work also confirms the existence of a quasi-sinusoidal oscillation in the timing noise of the Crab pulsar, that was reported by Richards et al. (1970), but with a smaller period, and with its amplitude and period decreasing with time, like a chirp signal. Such a coherent oscillation has not been noticed so far in either the Crab or any other pulsar. This work provides an explanation for the post-glitch behavior of the Crab pulsar glitches of 1969 Sep and 2004 Nov, and similar glitches in other pulsars, in terms of the recently proposed non-glitch speed-up event. If true, then non-glitch speed-up events may not be as rare as believed earlier. This work argues that it is unlikely that the frequency and amplitude modulated sinusoidal variation in the timing noise is due to unmodeled planetary companions.

Context: The rotation history of the Crab Pulsar is well described by (1) a rotation frequency $\nu$ and a slowdown model that is specified by its first two time derivatives $\dot \nu$ and $\ddot \nu$, known as the secular slowdown model, (2) occasional (once in $\approx 2$ years) significant and abrupt increases in the magnitude of $\nu$ and $\dot \nu$ (occurring on timescales of minutes), known as glitches, and (3) much slower increases and decreases in $\nu$ and $\dot \nu$ (occurring over months and years) that are an order of magnitude smaller, known as timing noise. Aims: This work reports a speed-up event in the Crab Pulsar that occurred around $2015$ February that is distinct from glitches and timing noise. Methods: Monthly $\nu$s and $\dot \nu$s of the Crab Pulsar, obtained at radio frequencies and published by Jodrell Bank Observatory (JBO), are used to demonstrate the speed-up event. Monthly arrival times of the Crab Pulsar's pulse, also published by JBO, combined with X-ray data from the RXTE, SWIFT, and NUSTAR observatories are used to verify the result. Results: The speed-up event is caused by a persistent increase in $\dot \nu$, which results in a monotonic increase in $\nu$. Over the last $\approx 550$ days, $\nu$ has increased monotonically by an amount that is $\approx 10$ times larger than the timing noise level. Conclusions: This is a unique event in the Crab Pulsar. This is probably due to a small increase in the Crab Pulsar's internal temperature. In its absence, the next large glitch in the Crab Pulsar is expected to occur around $2019$ March. However, this event could have an important bearing on its occurrence.

On 2006 August 23 the Crab Pulsar underwent a glitch, that was reported by the Jodrell Bank and the Xinjiang radio observatories. Neither data are available to the public. However, the Jodrell group publishes monthly arrival times of the Crab Pulsar pulse (their actual observations are done daily), using which it is shown that about five months earlier, the Crab Pulsar most probably underwent a small glitch, which has not been reported before. Neither observatory discusses the detailed analysis of data from 2006 March to August; either they may not have detected this small glitch, or may have attributed it to timing noise in the Crab Pulsar. The above result is verified using X-ray data from the RXTE observatory. If this is indeed true, this may probably be the smallest glitch observed in the Crab Pulsar so far, whose implications are discussed. This work addresses the confusion possible between small magnitude glitches and timing noise in pulsars.

In October 1997, radio pulses from the Crab Pulsar underwent abnormal delay. This was reported by two radio observatories, both of which explained this frequency dependent and time varying delay as being due to refractive effects of ionized shells in the Crab Nebula. Both groups also noted that, curiously and confusingly coincident with the frequency dependent delay, the Crab Pulsar also underwent an unusual slowing down, which they believed to be unrelated to the Crab Nebula and instead intrinsic to the Crab Pulsar, resulting in an additional delay that was frequency independent. However, it now appears that one of the groups attributes the frequency independent delay also to refractive effects. This work aims to verify whether at least a part of the frequency independent delay is indeed due to intrinsic slowing down of the Crab Pulsar. Timing analysis of the Crab Pulsar's October 1997 event has been done in X-rays, which are not delayed by the refractive and diffractive effects that affect radio waves; at X-rays only the intrinsic slowing down should contribute to any observed delay. Data mainly from the PCA instrument aboard the RXTE satellite have been used, along with a small amount of data from the PDS instrument aboard the BeppoSAX satellite. Analysis of the X-ray data, using the very accurate reference timing mode derived at radio frequencies, strongly supports the intrinsic slowing down hypothesis. Analysis using the reference timing model derived self-consistently from the limited X-ray data, which is less accurate, is not completely unambiguous regarding the above two hypotheses, but provides reasonable support for the intrinsic slowing down hypothesis. A significant fraction of the frequency independent delay during the October 1997 event is indeed due to intrinsic slowing down of the Crab Pulsar.

Hard xray data from the RXTE observatory (HEXTE energy range 15 to 240 keV) have been analyzed to obtain a phase coherent timing solution for the Crab pulsar glitch of 15 July 2000. The results are: (1) step change in the rotation frequency $\nu_0$ of the Crab pulsar at the epoch of the glitch is $\Delta \nu_0 = (30 \pm 3) \times 10^{-9} \times \nu_0$, (2) step change in its time derivative is $\Delta \dot \nu_0 = (4.8 \pm 0.6) \times 10^{-3} \times \dot \nu_0$, and (3) the time scale of decay of the the step change is $\tau_d = 4.7 \pm 0.5$ days. The first two results are consistent with those obtained at radio frequencies by the Jodrell Bank observatory. The last result has not been quoted in the literature, but could be an underestimate due to lack of observations very close to the glitch epoch. By comparing with the monthly timing ephemeris published by the Jodrell group for the Crab pulsar, the time delay between the main peaks of the hard xray and radio pulse profiles is estimated to be $+411 \pm 167$ $\mu$sec. Although this number is not very significant, it is consistent with the number derived for the 2 to 16 keV energy range, using the PCA instrument of RXTE. The separation between the two peaks of the integrated pulse profile of the Crab pulsar, and the ratio of their intensities, both are statistically similar before and after the glitch. The dead time corrected integrated photon flux within the integrated pulse profile appears to decrease after the glitch, although this is not a statistically strong result. This work achieves what can be considered to be almost absolute timing analysis of the Crab pulsar hard xray data.

Deshpande and Rankin (DR1999, DR2001) claim that the frequency of the very narrow feature, in the spectrum of radio flux variations of PSR B0943+10, is an alias of its actual value. For this, they need to interpret the narrow feature as a very stable pattern of drifting sub pulses on the polar cap of this pulsar, within the framework of the Ruderman and Sutherland (1975) model. This paper discusses several factors within the same framework, that may perturb a very steady pattern of drifting sub pulses, leading to a broadening of the spectral feature. One should be cautious in (1) identifying such very narrow spectral features with a very stable pattern of drifting sub pulses, and (2) use this picture to resolve the issue of aliasing in PSR B0943+10.

Deshpande and Rankin (DR1999, DR2001) claim that the frequency of the very narrow feature, in the spectrum of radio flux variations of PSR B0943+10, is an alias of its actual value. They also claim to have detected an amplitude modulation on the above phase modulation. This paper argues that both these claims are unjustified.

Pravdo et al (1997) claimed that the phase resolved x-ray spectrum in Crab pulsar (PSR B0531+21) shows a spectral hardening at the leading edge of the first peak of its integrated profile; this was a new and unexpected result. This article reanalyzes their data, as well as some other related data, and argues that the spectrum is as likely to be unvarying (i.e., neither hardening nor softening).

This article is the second in the series that analyze about 1.87 million periods of Crab pulsar, observed by the PCA detector aboard the RXTE x-ray observatory. At these energies the pulsar displays none of the three phenomena that are often seen in normal radio pulsars -- ``pulse nulling'', ``systematic sub pulse drifting'' and ``mode changing''. Presence or absence of these three behavior in the Crab pulsar, at radio wavelengths, something that has not been rigorously established yet, might be important for a satisfactory understanding of the above three phenomena.

Seven giant radio pulses were recorded from the millisecond pulsar PSR B1937+21 during $\approx$ 8.1 minutes observation by the Ooty Radio Telescope (ORT) at 326.5 MHz. Although sparse, these observations support most of the giant pulse behavior reported at higher radio frequencies (430 to 2380 MHz). Within the main componentof the integrated profile, they are emitted only in a narrow ($\la$ 47 $\mu$s) window of pulse phase, close to its peak. This has important implications for doing super-high precision timing of PSR B1937+21 at low radio frequencies.

The instability in one of the components of the radio integrated profile (IP) of the millisecond pulsar PSr J0437-4715 is correlated with the occurrence of the spiky emission in that component. It is speculated that the instability in other components is also due to the spiky emission. This might have important implications for interpreting the individual components of the IP as independent beams or regions of emission.

An independent analysis of the 326.5 MHz data obtained by Ooty Radio Telescope reveals no evidence for coherent diffractive radiation patterns in the millisecond pulsar PSR J0437-4715.

This article is the first in the series that analyze about 1.87 million periods of PSR B0531+21 (Crab pulsar), observed by the PCA detector aboard the RXTE x-ray observatory. The Crab pulsar's x-ray light curve shows little variation over time scales ranging from days to a period (33.46 milli seconds). The standard deviation of its x-ray flux variation is $\approx$ 0.7% of its mean value, which is negligible compared to its radio flux variations. The phase resolved power spectrum of pulse to pulse x-ray flux variation shows no spectral feature; an upper limit to the peak of any possible broad spectral feature is 0.06% of the mean power. The x-ray fluxes in the two components of its integrated profile are unrelated to each other; their linear correlation coefficient is 0.0004$\pm$0.0010. ``Giant pulses'' that are routinely seen at radio wavelengths are absent here. This work sets very strong constraints on the connection (if any) between the flux variations at radio and x-ray energies, for example due to variation in the degree of coherence of the basic emitters. Its phase resolved x-ray flux variation shows a weak correlation with the integrated profile. If confirmed, this might be an important clue to understanding the x-ray emission mechanism of Crab pulsar.

Time-scales of radio emission are studied in PSR J0437-4715 at 327 MHz using almost half a million periods of high quality data from Ooty Radio Telescope. The radio emission in this milli second pulsar occurs on a short (s) time-scale of approximately 0.026 +- 0.001 periods, and on a (l) time-scale that is much longer than the widths of the components of the integrated profile (approximately 0.05 periods). The width of the s emission increases with its increasing relative contribution to the total radio emission. This may provide constraints for the details of discharge of vacuum gaps above pulsar polar caps. The s emission occasionally takes place in the form of intense spikes, which are confined to the main component of the integrated profile for 90 per cent of the time. The positions of spikes within a component of the integrated profile have no simple relation to the shape of that component. This may have impact on the interpretation of the integrated profile components in terms of independent regions of emission on the polar cap.

The pair of drifting subpulses of the radio pulsar PSR B0031-07 appear to be well separated from each other on the average, overlapping at less than one thirtieth of the peak energy. Strictly, the integrated profile of the pair (after removing the drift) does not show more than two subpulses; however this result is likely to be modified with the availability of more data. On the other hand, rare instances of three simultaneous subpulses in a single period have been noticed. This is consistent with the Ruderman and Sutherland model of the drifting phenomenon in which several sparks, uniformly spaced on a circle, rotate around the polar cap. The third subpulse appears to show up strongly when subpulses belonging to the primary drift bands weaken. Although this conclusion is based mainly on two reliable examples, it suggests an anti correlation of the subpulse energies. The normalised correlation coefficient between the energies of the two subpulses in a period is -0.16 +- 0.02, after accounting for a known selection effect. Simulations suggest that this result could not be an artefact. This suggests the idea of competing drifting subpulses which, if true, will have important ramifications for the growth and breakdown of the particle acceleration zones in pulsars.