Search

The transient X-ray binary pulsar A0535+262 was observed with Suzaku on 2005 September 14 when the source was in the declining phase of the August-September minor outburst. The ~103 s X-ray pulse profile was strongly energy dependent, a double peaked profile at soft X-ray energy band (<3 keV) and a single peaked smooth profile at hard X-rays. The width of the primary dip is found to be increasing with energy. The broad-band energy spectrum of the pulsar is well described with a Negative and Positive power-law with EXponential (NPEX) continuum model along with a blackbody component for soft excess. A weak iron K_alpha emission line with an equivalent width ~25 eV was detected in the source spectrum. The blackbody component is found to be pulsating over the pulse phase implying the accretion column and/or the inner edge of the accretion disk may be the possible emission site of the soft excess in A0535+262. The higher value of the column density is believed to be the cause of the secondary dip at the soft X-ray energy band. The iron line equivalent width is found to be constant (within errors) over the pulse phase. However, a sinusoidal type of flux variation of iron emission line, in phase with the hard X-ray flux suggests that the inner accretion disk is the possible emission region of the iron fluorescence line., Comment: 21 pages, 6 figures. Accepted for publication in the Astrophysical Journal, 2008 January issue

The binary X-ray pulsar A0535+262 was observed with the Suzaku X-ray observatory, on 2005 September 14 for a net exposure of 22 ksec. The source was in a declining phase of a minor outburst, exhibiting 3--50 keV luminosity of about $3.7 \times 10^{35}$ ergs s$^{-1}$ at an assumed distance of 2 kpc. In spite of the very low source intensity (about 30 mCrab at 20 keV), its electron cyclotron resonance was detected clearly with the Suzaku Hard X-ray Detector, in absorption at about 45 keV. The resonance energy is found to be essentially the same as those measured when the source is almost two orders of magnitude more luminous. These results are compared with the luminosity-dependent changes in the cyclotron resonance energy, observed from 4U 0115+63 and X 0331+53., Comment: 12 pages, 4 figures, accepted to ApJL, high resolution version is available from http://cosmic.riken.jp/terada/01work/paper/a0535_suzaku2006.pdf

We present here high resolution X-ray spectrum of the accreting binary X-ray pulsar GX 1+4 obtained with the High Energy Transmission Grating (HETG) instrument of the Chandra X-ray Observatory. This was supplemented by a simultaneous observation with the Proportional Counter Array (PCA) of the Rossi X-ray Timing Explorer (RXTE). During this observation, the source was in a somewhat low intensity state and the pulse profile with both Chandra and RXTE shows a narrow dip, characteristic of GX 1+4 in medium and low intensity states. The continuum X-ray spectrum obtained with the HETG and PCA can be fitted well with a high energy cutoff power-law model with line of sight absorption. Interestingly, we find that this low state is accompanied by a relatively small absorption column density. A 6.4 keV narrow emission line with an equivalent width of 70 eV is clearly detected in the HETG spectrum. The fluorescence iron line, or at least part of it is produced in the neutral or lowly ionized iron in the circumstellar material that also causes most of the line of sight absorption. In the HETG spectrum, we have found evidence for a weak (equivalent width ~30 eV) emission line at 6.95 keV. This line is identified as Ly_alpha emission line from hydrogen-like iron and the spectrum does not show emission lines from helium-like iron. We discuss various emission regions for the hydrogen-like iron emission line, like gas diffused into the Alfven sphere or an accretion curtain flowing from the inner accretion disk to the magnetic poles., Comment: 15 pages, 4 postscript figures, accepted for publication in The Astrophysical Journal

We present an INTEGRAL observation of the Cen-Crux region in order to search the electron cyclotron resonance scattering features from the X-ray binary pulsars. During the AO1 200ks observation, we clearly detected 4 bright X-ray binaries, 1 Seyfert Galaxy, and 4 new sources in the field of view. Especially from GX301-2, the cyclotron resonance feature is detected at about 37 keV, and width of 3--4 keV. In addition, the depth of the resonance feature strongly depends on the X-ray luminosity. This is the first detection of luminosity dependence of the resonance depth. The cyclotron resonance feature is marginally detected from 1E1145.1-6141. Cen X-3 was very dim during the observation and poor statistics disable us to detect the resonance features.These are first INTEGRAL results of searching for the cyclotron resonance feature., Comment: 4pages, 8figures, To be published in the Proceedings of the 5th INTEGRAL Workshop: "The INTEGRAL Universe", February 16-20, 2004, Munich

A review of spectroscopic results obtained from Chandra High Energy Transmission Grating Spectrometer and XMM-Newton Reflection Grating Spectrometer observations of several wind-fed high-mass X-ray binaries (HMXBs) is presented. These observations allow us to study the structure of the stellar wind in more detail and provide, for the first time, a dyanmical view of the X-ray photoionized wind that surrounds the compact object. At the same time, however, they are also providing us with numerous puzzles that cannot be explained in terms of simple models. For example, simple spherically-symmetric wind models cannot explain the observed orbital-phase variability of the line intensities and shapes, which may be caused by intrinsic asymmetries due to the presence of the compact object and/or more complicated radiative transfer effects. The observed line shifts are smaller than those expected from extensions of simple wind models of isolated OB supergiants. In addition, several novel spectroscopic discoveries have been made, including: (1) P-Cygni lines from an expanding wind, (2) detection of multiple Si K fluorescent lines from a wide range of charge states, (3) Compton scattered Fe K lines from a cold medium. We discuss how these spectroscopic diagnostics can be used to understand some of the global properties of stellar winds in HMXBs., Comment: 10 pages; invited review at the High-resolution X-ray Spectroscopy Workshop with XMM-Newton and Chandra, MSSL, Oct 24-25, 2002

We present results from ultraviolet and X-ray observations of the low mass X-ray binary Cygnus X-2. The simultaneous HST/GHRS and ASCA observations took place during the low state of an 82-day cycle. We compare our observations as well as archival IUE and RXTE data with models that predict ultraviolet and optical continuum emission from an X-ray heated disk and a Roche-lobe-filling star. The model predictions are consistent with observed optical, ultraviolet, and X-ray variations over both orbital and long-term periods. The X-ray spectral state, the luminosities implied by fits to the X-ray data, the ultraviolet continuum and line fluxes, and the mass accretion rates obtained from fits to the ultraviolet continuum are consistent with location of our observations on the normal and horizontal branches of the Z-shaped X-ray color-color diagram. A combination of changes to mass accretion rate and obstruction by a warped disk can be invoked as a possible explanation for the motion of the ``Z'' in the color-color plane. The GHRS/G160M measurements concentrated on NV and HeII. The low-resolution (GHRS/G140L) observations captured SiIV, NIV, and CIV. Although the relative line fluxes are consistent with emission from an X-ray heated accretion disk corona, predictions from models of line emission from simple disks do not fit the observed emission line profiles.The radial velocities (80-130 km s$^{-1}$) are consistent with emission from the optical star at the orbital phase (0.70-0.74) of our observations., Comment: 35 pages, 11 figures. Accepted for Sept. 2003 issue of PASP

We present here the results of an investigation of the pulse averaged and pulse phase resolved energy spectra of two high luminosity accretion powered X-ray pulsars SMC X-1 and LMC X-4 made with ASCA. The phase averaged energy spectra definitely show the presence of a soft excess in both the sources. If the soft excess is modeled as a separate black-body or thermal bremsstrahlung type component, pulse phase resolved spectroscopy of SMC X-1 shows that the soft component also has a pulsating nature. Same may be true for LMC X-4, though a very small pulse fraction limits the statistical significance. The pulsating soft component is found to have a nearly sinusoidal profile, dissimilar to the complex profile seen at higher energies, which can be an effect of smearing. Due to very high luminosity of these sources, the size of the emission zone required for the soft component is large (radius ~300-400 km). We show that the pulsating nature of the soft component is difficult to explain if a thermal origin is assumed for it. We further investigated with alternate models, like inversely broken power-law or two different power-law components and found that these models can also be used to explain the excess at low energy. A soft power-law component may be a common feature of the accreting X-ray pulsars, which is difficult to detect because most of the HMXB pulsars are in the Galactic plane and experience large interstellar absorption. In LMC X-4, we have also measured two additonal mid-eclipse times, which confirm the known orbital decay., Comment: 28 pages (AAS preprint), 10 Postscript figures, ApJ accepted

ASCA's long look at the 288 millisecond radio pulsar, PSR J0631+1036, reveals coherent X-ray pulsation from this source for the first time. The source was first detected in the serendipitous Einstein observation and later identified as a radio pulsar. Possible pulsation in the gamma-ray band has been detected from the CGRO EGRET data (Zepka, et al. 1996). The X-ray spectrum in the ASCA band is characterized by a hard power-law type emission with a photon index of about 2.3, when fitted with a single power-law function modified with absorption. An additional blackbody component of about 0.14 keV increases the quality of the spectral fit. The observed X-ray flux is 2.1e-13 ergs/s/cm2 in the 1-10 keV band. We find that many characteristics of PSR J0631+1036 are similar to those of middle-aged gamma-ray pulsars such as PSR B1055-52, PSR B0633+17 (Geminga), and PSR B0656+14., Comment: To appear in ApJ Letters

We report observation results of the super-soft X-ray sources (SSS) CAL87 and RXJ0925.7-4758 with the X-ray CCD cameras (Solid-state Imaging Spectrometer; SIS) on-board the ASCA satellite. We have applied theoretical spectral models to CAL87, and constrained the white dwarf mass and intrinsic luminosity as 0.8 - 1.2 M_solar and 4 x 10^{37}- 1.2 x 10^{38} erg s^{-1}, respectively. However, we have found the observed luminosity is an order of magnitude smaller than the theoretical estimate, which indicates the white dwarf is permanently blocked by the accretion disk, and we are observing a scattering emission by a fully ionized accretion disk corona (ADC) whose column density is ~ 1.5 x 10^{23} cm^{-2}. Through simulation, we have shown that the orbital eclipse can be explained by the ADC model, such that a part of the extended X-ray emission from the ADC is blocked by the companion star filling its Roche lobe. We have found that very high surface gravity and temperature, ~ 10^{10} cm s^{-2} and ~ 100 eV respectively, as well as a strong absorption edge at ~ 1.02 keV, are required to explain the X-ray energy spectrum of RXJ0925.7-4758. These values are only possible for an extremely heavy white dwarf near the Chandrasekhar limit. Although the super soft source luminosity should be ~ 10^{38} erg s^{-1} at the Chandrasekhar limit, the observed luminosity of RXJ0925.7-4758 is nearly two orders of magnitude smaller even assuming an extreme distance of ~10 kpc. To explain the luminosity discrepancy, we propose a model that very thick matter which was previously ejected from the system, as a form of jets, intervenes the line of sight, and reduces the luminosity significantly due to Thomson scattering., Comment: ApJ accepted

XTE J1748-288 is a new X-ray transient with a one-sided radio jet. It was observed with ASCA on 1998/09/06 and 1998/09/26, 100 days after the onset of the radio-X-ray outburst. The spectra were fitted with an attenuated power-law model, and the 2-6-keV flux was 4.6 * 10^{-11} erg s^{-1} cm^{-2} and 2.2 * 10^{-12} on 09/06 and 09/26, respectively. The light curve showed that the steady exponential decay with an e-folding time of 14 days lasted over 100 days and 4 orders of magnitude from the peak of the outburst. The celestial region including the source had been observed with ASCA on 1993/10/01 and 1994/09/22, years before the discovery. In those period, the flux was < 10^{-13} erg s^{-1} cm^{-2}, below ASCA's detection limit. The jet blob colliding to the environmental matter was supposedly not the X-ray source, although the emission mechanism has not been determined. A possible detection of a K line from highly ionized iron is discussed., Comment: 11 pages, 4 figures, submitted to ApJL. Fig2 is replaced with correct one

We present new observations of two anomalous X-ray pulsars (AXP) 4U 0142+61 and 1E 1048.1-5937 made in 1998 with the ASCA. The energy spectra of these two AXPs are found to consist of two components, a power-law and a blackbody emission from the neutron star surface. These observations, when compared to earlier ASCA observations in 1994 show remarkable stability in the intensity, spectral shape and pulse profile. However, we find that the spin-down rate in 1E 1048.1-5937 is not constant. In this source, we have clearly identified three epochs with spin-down rates different from each other and the average value. This has very strong implications for the magnetar hypothesis of AXPs. We also note that the spin-down rate and its variations in 1E 1048.1-5937 are much larger than what can normally be produced by an accretion disk with very low mass accretion rate corresponding to its low X-ray luminosity., Comment: 12 pages, 7 postscript figures included, To appear in the Astrophysical Journal

We present results of a comprehensive study of the distribution of absorption dips with orbital phase in Cygnus X-1. Firstly, the distribution was obtained using archival data from all major X-ray observatories and corrected for the selection effect that phase zero (superior conjunction of the black hole) has been preferentially observed. Dip occurrence was seen to vary strongly with orbital phase \phi, with a peak at \phi ~ 0.95, i.e. was not symmetric about phase zero. Secondly, the RXTE ASM has provided continuous coverage of the Low State of Cygnus X-1 since Sept. 1996, and we have selected dip data based on increases in hardness ratio. The distribution, with much increased numbers of dip events, confirms that the peak is at \phi ~ 0.95, and we report the discovery of a second peak at \phi ~ 0.6. We attribute this peak to absorption in an accretion stream from the companion star HDE 226868. We have estimated the ionization parameter at different positions showing that radiative acceleration of the wind is suppressed by photoionization in particular regions in the binary system. To obtain the variation of column density with phase, we make estimates of neutral wind density for the extreme cases that acceleration of the wind is totally suppressed, or not suppressed at all. An accurate description will lie between these extremes. In each case, a strong variation of column density with orbital phase resulted, similar to the variation of dip occurrence. This provides evidence that formation of the blobs in the wind which lead to absorption dips depends on the density of the neutral component in the wind, suggesting possible mechanisms for blob growth., Comment: 9 pages, Latex, 7 ps figures. accepted by MNRAS

We observed the HH24-26 region in the L1630 Orion molecular cloud complex with the X-ray observatory ASCA in the 0.5$-$10 keV band. X-ray emission was detected from the T Tauri star SSV61 and from the region where the Class I protostars SSV63E and SSV63W are located (hereafter SSV63E+W). The spectra of both SSV63E+W and SSV61 are well explained by an optically thin thermal plasma model. The spectrum of the T Tauri star SSV61 has a low temperature of $kT=0.9$ (0.7$-$1.2) keV and a moderate absorption of $N_{\rm{H}}=1.3$ (0.9$-$1.7) $\times10^{22}$ cm$^{-2}$, while that of the protostar SSV63E+W has a high temperature of $kT=5.0$ (3.3$-$7.9) keV and a heavy absorption of $N_{\rm{H}}=1.5$ (1.2$-$1.8) $\times10^{23}$ cm$^{-2}$. The X-ray light curve of SSV63E+W showed a flare during the observation. The peak flux reached about 9 times that of the quiescent flux. The temperature and the absorption column density do not change conspicuously during the flare. The 0.5$-$10 keV luminosity of SSV63E+W was about $1\times10^{32}$ erg s$^{-1}$ in the quiescent state. The present detection of hard X-rays from SSV63E+W is remarkable, because this is the first X-ray detection of a protostar in Orion., Comment: 14 pages, 3 postscript figures, uses aasms4.sty

We report the results from X-ray and hard X-ray observations of the PSR B1259-63/SS 2883 system with the ASCA and CGRO satellites, performed between 1995 February and 1996 January when the pulsar was near apastron. The system was clearly detected in each of the two ASCA observations with luminosity in the 1-10 keV band of Lx = (9 +- 3) x10^32 (d/2 kpc)^2 erg/s, while CGRO/OSSE did not detect significant hard X-rays from the system. X-ray spectra obtained with ASCA are well-fit by a single power-law spectrum with a photon index of 1.6 +- 0.3. No pulsations were detected in either the ASCA or the OSSE data. We combine all existing X-ray and hard X-ray observations and present the orbital modulation in the luminosity and photon index for the entire orbit. The results are in agreement with predictions based on a synchrotron emission model from relativistic particles in a shocked pulsar wind interacting with the gaseous outflow from the Be star., Comment: 14 pages, 4 figures, accepted for publicatoin in ApJ

We report the ASCA results of the bursting X-ray pulsar \gro, which was observed in February 1996 and March 1997. The source flux in the 2--10 keV band was $2.0\times10^{-8}$ erg/sec/cm$^2$ in 1996 and $5.0\times10^{-9}$ erg/sec/cm$^2$ in 1997. We detected 12 and 17 Type II bursts during the two observations with mean bursting intervals of about 27 min and 37 min. Each burst is followed by an intensity dip with the depleted flux depending on the burst fluence. The energy spectra are approximated by an absorbed power law with additional structure around 6--7 keV\@. Constant absorption column, $(5-6)\times10^{22}$ cm$^{-2}$, independent of the observation dates and emission phases (persistent, burst and dip) is interpreted as an interstellar absorption. The source may be actually located near the Galactic center, at a distance of 8.5 kpc. The structure in the energy spectrum at 6--7 keV is most probably due to iron and maybe reproduced by a disk line model with additional broadening mechanism., Comment: 28 pages 11 figures. ApJ accepted

We present first results from simultaneous ultraviolet (UV; HST/GHRS) and X-ray (ASCA) observations of the 13.5s pulsar LMC X-4 taken in 1996 May. The GHRS data are the highest resolution (both temporal and spectral) UV spectra ever taken of LMC X-4. Fits to the UV continuum using a model that incorporates X-ray heating of the companion star and the accretion disk yields a mass accretion rate Mdot=4.0x10^(-8) Msun/yr; the X-ray luminosity implied by this value is consistent with the X-ray flux measured during simultaneous observations. The model accurately predicts observed B magnitude and ultraviolet variations over both orbital and long-term periods. The UV P-Cygni lines show dramatic changes with orbital phase: strong broad absorption near X-ray eclipse and narrow absorption when the X-ray source is in the line-of-sight. We interpret this as a result of X-ray photoionization of the stellar wind; when the neutron star is in front of the normal star, the wind absorption disappears and mainly the photospheric absorption lines are visible. The X-ray pulse period measured during our observations, 13.5090+/-0.0002s, is consistent with steady spin-down over the past 10 years. No pulsations were detected in the UV observations with upper limits to the pulsed fraction around N V and C IV of 1.8% and 2.7% in the continuum and 12.4% and 7% in the absorption troughs., Comment: 21 pages, including 5 figures, submitted to the Astrophysical Journal, uses aaspp4.sty, psfig

We report on ASCA energy spectra of Cir X-1 taken near its zero phase on 1994 August 4-5. The ASCA SIS detectors allow a more detailed study of the iron K complex than has been possible before. We find that prior to a sudden upward flux transition the dominant iron K feature appears to consist of a large edge from neutral or nearly-neutral iron. The depth of the edge corresponds to an absorption column of about 1.5E24 cm^{-2}, while little absorption over that expected from the Galaxy is seen at lower X-ray energies. The differential absorption at high and low X-ray energies combined with the iron edge energy are strong evidence that partial covering is a crucial determinant of the behaviour observed from Cir X-1. The continuum spectral variability observed by ASCA can also be understood naturally in terms of partial covering column changes. There is evidence for a relatively weak emission line from neutral iron with an equivalent width of only about 65 eV. After the flux transition, the strength of the edge feature is greatly reduced, suggesting a large reduction in the amount of partial covering. For a large region of statistically acceptable chi-squared parameter space, the luminosity of Cir X-1, after correction for partial covering, need not change during the transition. We discuss models for the partial covering and suggest that X-ray scattering by electrons may be important. Aspects of the Cir X-1 spectrum are very similar to those of Seyfert 2 galaxies with Compton-thin tori., Comment: 14 pages, uuencoded compressed tar file, MNRAS

We present eclipse timing measurements of the low-mass X-ray binary (LMxB) EXO 0748-676 made with ASCA. These measurements, when combined with previous results, are consistent with the hypothesis that eclipse times are modulated by a third body. The parameters of the orbit of the postulated third body are now apparently well constrained and will allow further observations to effectively test the hypothesis that EXO 0748-676 is a triple system. The derived orbital parameters are such that, unless the inclination angle is very low, a third body in the system must have a mass less than the lower limit for a main-sequence star.

We report a 1.7 day ASCA X-ray observation of the 2.87 day binary Algol (Beta Per), centered on the secondary eclipse. Spectra accumulated for different intensity states show a prominent He-like iron K line at 6.7 keV. A two-temperature variable abundance plasma model applied to the spectra yielded temperatures of approximately 8 and approximately 30 MK. The modeled coronal abundances of Fe, O, Mg, Si, S, Ar, and Ca were a factor of 2-3 below the solar photospheric value, and N less than 0.1. These model abundance anomalies are similar to those found from the ASCA spectra of other late-type stars and may indicate either true deviations from solar abundances or problems with the assumptions and atomic physics of the plasma models. The X-ray light curve shows a factor of 2 increase in flux over 13 hr beginning in the middle of the secondary eclipse, with a shallow eclipse centered on phase 0.45. The orbital light curve is similar to that observed by ROSAT 1 year earlier. The rise in flux is caused by an increase in the emission measure of the higher temperature component. The intensity variation is not associated with any major change in the abundances or temperature of the two components.

We analyze data from two eclipse ingresses of Her X-1 observed with Ginga on 1989 April 30 and May 19. These observations occur, respectively, during the MAIN HIGH and SHORT HIGH states in the 35 day modulation of Her X-1 intensity. We find significant residual X-ray flux during eclipse, with a gradual decrease in flux following the occultation of the neutron star by the atmosphere of HZ Her. During the central part of the eclipse the count rate becomes nearly constant, at 0.5 mcrab in the energy range 1.7-36.8 keV. From a spec- tral analysis of the residual emission during the total eclipse of the central source in the MAIN MGH state, we determine the energy spectral index, alpha = 0.8, similar to that before eclipse. A remarkable feature of the eclipse spectrum is that it does not show a significant iron line feature in contrast to massive wind-fed pulsars, such as Vela X-1 and Cen X-3. From a timing analysis of the same eclipse data, we show that there are no pulses. These results imply that the emission comes from the scattering of continuum X-rays by material in a region considerably larger than the companion star. An extended accretion disk corona may be responsible for this scattering. However, partial eclipse of an extended accretion disk corona is insufficient to account for the count rates in mid-eclipse, when known parameters of the binary system are used. Based on the present results, we suggest that scattering occurs not only in the accretion disk corona but also in the circumstellar matter surrounding the system of Her X-1/HZ Her.

We report on the results of three observations of the binary X-ray pulsar SMC X-1 with the Ginga satellite. Timing analyses of the 0.71 s X-ray pulsations yield Doppler delay curves which, in turn, enable the most accurate determination of the SMC X-1 orbital parameters available to date. Epochs of phase zero for the 3.9 day orbit were determined for 1987 May, 1988 August, and 1989 August with accuracies of 13, 0.6, and 3 s, respectively. These epochs are combined with two previous determinations of the orbital epoch to yield the rate of change in the orbital period dot-P(orb)/P(orb) = ( 3.36 +/- 0.02) x 10(exp -6) yr(exp -1). An interpretation of the orbital decay is made in the context of tidal evolution, with consideration of the influence of the increasing moment of inertia of the companion star due to its nuclear evolution. We find that, while the orbital decay is probably driven by tidal interactions, the asynchronism between the orbit and the rotation of the companion star is most likely maintained by the evolutionary expansion of the companion star (Sk 160) rather than via the Darwin instability. In this case Sk 160 is likely to be in the hydrogen shell burning phase of its evolution. Finally, a discussion is presented of the relation among the time scales for stellar evolution (less than 10(exp 7) yr), orbital decay (3 x 10(exp 5) yr), and neutron-star spin-up in the SMC X-1 system (2000 yr). In particular, we present the result of a self-consistent calculation for the histories of the spin of the neutron star and the mass transfer in this system. A plausible case can be made for the spin-up time scale being directly related to the lifetime of the luminous X-ray phase which will end in a common-envelope phase within a time of less than approx. 10(exp 4) yr.

We present the first discovery of pulsed iron line emission from an X-ray binary, namely Cen X-3. Compared with the continuum pulsations, the iron line pulsations are shallow (50 percent change in amplitude), smeared (the profile is a single-peaked sinusoid) and phase-shifted (by about half a cycle). We also discuss the constraints on the origin of the line imposed by this discovery and by other observations.

The Ginga X-ray satellite observed Algol (Beta Per) for 2 days in 1989 January, including both the primary optical eclipse and most of the secondary eclipse. We derive upper limits of about 20 and 10 percent, respectively, for the eclipsed flux fraction during the two eclipses. A large flare lasting over 12 hr was seen prior to and during secondary eclipse. High-temperature Fe line emission is clearly detected in the proportional counter data. The Fe line equivalent width is variable during the flare, ranging from 0.4-1.0 keV. Except for two intervals during the flare rise, the observed equivalent width is lower than predicted using solar abundances and an optically thin plasma model. Similar behavior has also been observed by Ginga in a large flare on UX Ari: in both events, opacity effects at line center may be playing a significant role. Loop model analysis of the large flare suggests that it involves a substantially longer loop or loops than a shorter duration Algol flare seen with Exosat.

Attention is given to a series of observations of Centaurus X-3 made with Ginga on March 22-24, 1989, over a complete orbital cycle. The data set includes a preeclipse dip, the eclipse ingress and egress, the eclipse itself, and a phase of high steady emission after egress. The mideclipse ephemeris obtained yields an estimate of the rate of change of the orbital period of -(1.738 +/- 0.004) x 10 exp 6/yr, confirming the rate of change reported by Kelley et al. (1983) and improving the precision by a factor of 20. The pulse profiles obtained with Ginga during the high postegress phase are double-peaked at low energies, and show a reversal of phase of 180 deg. No pulsations are detected during the eclipse and at low energies during the preeclipse dip, indicating that X-ray emission at these phases is due to scattering by extended matter. The energy of the iron emission line varies between 6.5 and 6.7 keV during the observation, while its equivalent width varies between 0.18 and 1.5 keV.

Spectral data from the binary X-ray pulsar 4 U 0900 - 40 obtained with the Ginga satellite and covering a range of orbital phases are presented and interpreted using simulated spectra created with a Monte Carlo scattering code. It is found that scattering and fluorescence in a simple spherically symmetric distribution of matter surrounding the companion star can reproduce the eclipse spectrum, as well as the 'soft X-ray-excess' observed during egress and other orbital phases. Reasonably secure values are found for a number of the parameters that characterize the density profile of the stellar wind and atmosphere of the companion star. The Fe abundance is within a factor of about 1.3 of that in the solar neighborhood. It is shown that ionization zones are not critical to understanding the orbital-phase-resolved spectra in this source. It is also found that the contribution by scattering from interstellar dust grains to the observed spectra during eclipse is negligible, while that from diffuse emission from the 'Galactic ridge' is significant.

Algol was observed by the Ginga X-ray satellite for about 2 days in January 1989, including both the primary and most of the secondary optical eclipses. No evidence for X-ray eclipses was seen. A large flare lasting over 12 h was detected prior to and during secondary eclipse. The flare began with a peak temperature of about 69 MK, gradually decaying to about 36 MK. Variable Fe line emission at 6.7 keV ws observed. The Fe line equivalent width is in clear disagreement with theoretical predictions for an optically thin plasma.

The results of a search for correlations between X-ray flux and angular acceleration for the accreting binary pulsar Vela X-1 are presented. Results are based on data obtained with the Hakucho satellite during the interval 1982 to 1984. In undertaking this correlation analysis, it was necessary to modify the usual statistical method to deal with conditions imposed by generally unavoidable satellite observing constraints, most notably a mismatch in sampling between the two variables. The results are suggestive of a correlation between flux and the absolute value of the angular acceleration, at a significance level of 96 percent. The implications of the methods and results for future observations and analysis are discussed.

The extensive observations of Vela X-1 acquired from 1980 to 1984 with the Hakucho and Tenma satellites, using a pulse-filtering technique that provides a significant improvement in pulse-timing precision, are reexamined. The details of the preparation of this data base, including the determination of updated orbital parameters based on these and previously published observations covering a 10-yr interval, are reported. Preparatory to estimation of angular-acceleration flux-correlation pairs, substantially improved estimates of the power density of fluctuations in the derivative of the pulse frequency on analysis periods between 0.5 and 80 days are presented. These results support the phenomenological hypothesis that unresolved fluctuations in the angular acceleration of the neutron star, describable as random noise, are responsible for the entire history of the observed variation in pulse frequency of Vela X-1 on all time scales longer than a few days.

X-ray outbursts from GX 301-2 were observed with Hakucho on two occasions during April and May 1982 and with Tenma in April 1984. Pulse arrival times measured with Hakucho were combined with those previously measured with Ariel 5 and SAS 3. The joint timing analysis of these data sets yields an orbital period of 41.5 days as the only acceptable solution; this is consistent with the period derived from the X-ray flaring events of the source. The orbital elements of the binary system, established from the joint timing analysis, are presented. The orbital solution suggests that all the peaks of X-ray flares observed during the past 10 yr occur at the orbital period, but consistently appear about 1.4 days before the time of periastron passage of the X-ray star.