A sub-set of the brightest ultraluminous X-ray sources (ULXs), with X-ray luminosities well above $10^{40}$ erg s$^{-1}$, typically have energy spectra which can be well described as hard power-laws, and short-term variability in excess of $\sim10\%$. This combination of properties suggests that these ULXs may be some of the best candidates to host intermediate mass black holes (IMBHs), which would be accreting at sub-Eddington rates in the hard state seen in Galactic X-ray binaries. In this work, we present a temporal and spectral analysis of all of the available XMM-Newton data from one such ULX, the previously poorly studied 2XMM J143242.1$-$440939, located in NGC 5643. We report that its high quality EPIC spectra can be better described by a broad, thermal component, such as an advection dominated disc or an optically thick Comptonising corona. In addition, we find a hint of a marginal change in the short-term variability which does not appear to be clearly related to the source unabsorbed luminosity. We discuss the implications of these results, excluding the possibility that the source may be host an IMBH in a low state, and favouring an interpretation in terms of super-Eddington accretion on to a black hole of stellar origin. The properties of NGC 5643 ULX1 allow us to associate this source to the population of the hard/ultraluminous ULX class.

With the advent of more sensitive all-sky instruments, the transient Universe is being probed in greater depth than ever before. Taking advantage of available resources, we have established a comprehensive database of black hole (and black hole candidate) X-ray binary (BHXB) activity between 1996 and 2015 as revealed by all-sky instruments, scanning surveys, and select narrow-field X-ray instruments aboard the INTErnational Gamma-Ray Astrophysics Laboratory (INTEGRAL), Monitor of All-Sky X-ray Image (MAXI), Rossi X-ray Timing Explorer (RXTE), and Swift telescopes; the Whole-sky Alberta Time-resolved Comprehensive black-Hole Database Of the Galaxy or WATCHDOG. Over the past two decades, we have detected 132 transient outbursts, tracked and classified behavior occurring in 47 transient and 10 persistently accreting BHs, and performed a statistical study on a number of outburst properties across the Galactic population. We find that outbursts undergone by BHXBs that do not reach the thermally dominant accretion state make up a substantial fraction ($\sim$ 40%) of the Galactic transient BHXB outburst sample over the past $\sim20$ years. Our findings suggest that this "hard-only" behavior, observed in transient and persistently accreting BHXBs, is neither a rare nor recent phenomenon and may be indicative of an underlying physical process, relatively common among binary BHs, involving the mass-transfer rate onto the BH remaining at a low level rather than increasing as the outburst evolves. We discuss how the larger number of these "hard-only" outbursts and detected outbursts in general have significant implications for both the luminosity function and mass-transfer history of the Galactic BHXB population.

The known members of the class of hyperluminous X-ray sources (HLXs) are few in number, yet they are of great interest as they are regarded as the likeliest intermediate-mass black hole (IMBH) candidates amongst the wider population of ultraluminous X-ray sources (ULXs). Here we report optical photometry and spectroscopy of a HLX candidate associated with the galaxy IC 4320, that reveal it is a background AGN. We discuss the implications of the exclusion of this object from the small number of well-studied HLXs, that appears to accentuate the difference in characteristics between the good IMBH candidate ESO 243-49 HLX-1 and the small handful of other HLXs.

We use K-band spectroscopy of the counterpart to the rapidly variable X-ray transient XMMU J174445.5-295044 to identify it as a new symbiotic X-ray binary. XMMU J174445.5-295044 has shown a hard X-ray spectrum (we verify its association with an Integral/IBIS 18-40 keV detection in 2013 using a short Swift/XRT observation), high and varying N$_H$, and rapid flares on timescales down to minutes, suggesting wind accretion onto a compact star. We observed its near-infrared counterpart using the Near-infrared Integral Field Spectrograph (NIFS) at Gemini-North, and classify the companion as ~ M2 III. We infer a distance of $3.1^{+1.8}_{-1.1}$ kpc (conservative 1-sigma errors), and therefore calculate that the observed X-ray luminosity (2-10 keV) has reached to at least 4$\times10^{34}$ erg/s. We therefore conclude that the source is a symbiotic X-ray binary containing a neutron star (or, less likely, black hole) accreting from the wind of a giant.

We present the first \textit{Chandra} and \textit{Swift} X-ray study of the spiral galaxy NGC\,4088 and its ultraluminous X-ray source (ULX N4088--X1). We also report very long baseline interferometry (VLBI) observations at 1.6 and 5 GHz performed quasi-simultaneously with the \textit{Swift} and \textit{Chandra} observations, respectively. Fifteen X-ray sources are detected by \textit{Chandra} within the D25 ellipse of NGC\,4088, from which we derive the X-ray luminosity function (XLF) of this galaxy. We find the XLF is very similar to those of star-forming galaxies and estimate a star-formation rate of 4.5 $M_{\odot}$ yr$^{-1}$. The \textit{Chandra} detection of the ULX yields its most accurate X-ray position, which is spatially coincident with compact radio emission at 1.6 GHz. The ULX \textit{Chandra} X-ray luminosity, $L_\mathrm{0.2-10.0 keV} = 3.4\ \times$ 10$^{39}$ erg s$^{-1}$, indicates that N4088--X1 could be located at the high-luminosity end of the high-mass X-ray binary (HMXB) population of NGC\,4088. The estimates of the black hole mass and ratio of radio to X-ray luminosity of N4088--X1 rule out a supermassive black hole nature. The \textit{Swift} X-ray spectrum of N4088--X1 is best described by a thermal Comptonization model and presents a statistically significant high-energy cut-off. We conclude that N4088--X1 is most likely a stellar remnant black hole in a HMXB, probably fed by Roche lobe overflow, residing in a super-Eddington ultraluminous state. The 1.6 GHz VLBI source is consistent with radio emission from possible ballistic jet ejections in this state.

We report and study the outburst of a new transient X-ray binary (XRB) in Terzan 5, the third detected in this globular cluster, Swift J174805.3-244637 or Terzan 5 X-3. We find clear spectral hardening in Swift/XRT data during the outburst rise to the hard state, thanks to our early coverage (starting at L_X ~ 4x10^{34} ergs/s) of the outburst. This hardening appears to be due to the decline in relative strength of a soft thermal component from the surface of the neutron star (NS) during the rise. We identify a {Type I X-ray burst} in Swift/XRT data with a long (16 s) decay time, indicative of {hydrogen burning on the surface of the} NS. We use Swift/BAT, Maxi/GSC, Chandra/ACIS, and Swift/XRT data to study the spectral changes during the outburst, identifying a clear hard-to-soft state transition. We use a Chandra/ACIS observation during outburst to identify the transient's position. Seven archival Chandra/ACIS observations show evidence for variations in Terzan 5 X-3's non-thermal component, but not the thermal component, during quiescence. The inferred long-term time-averaged mass accretion rate, from the quiescent thermal luminosity, suggests that if this outburst is typical and only slow cooling processes are active in the neutron star core, such outbursts should recur every ~10 years.

In this paper we present a combined analysis of data obtained with the Hubble Space Telescope (HST), Very Large Telescope (VLT), and Swift X-ray telescope (XRT) of the intermediate mass black hole ESO 243-49 HLX-1 that were taken 2 months apart between September and November 2010. Previous separate analyses of these data found that they were consistent with an irradiated accretion disc with contribution from either a very young or very old stellar population, and also indicated that the optical flux of the HLX-1 counterpart could be variable. Such variability could only be attributed to a varying accretion disc, so simultaneous analysis of all data sets should break the degeneracies in the model fits. We thus simultaneously fit the broad-band spectral energy distribution (SED) from near-infrared through to X-ray wavelengths of the two epochs of data with a model consisting of an irradiated accretion disc and a stellar population. We show that this combined analysis rules out an old stellar population, finding that the SED is dominated by emission from an accretion disc with moderate reprocessing in the outer disc around an intermediate mass black hole imbedded in a young (20 Myr) stellar cluster with a mass of 1E5 Msun. We also place an upper limit on the mass of an additional hidden old stellar population of 1E6 Msun. However, optical r-band observations of HLX-1 obtained with the Gemini-South telescope covering part of the decay from a later X-ray outburst are consistent with constant optical flux, indicating that the observed variability between the HST and VLT observations could be spurious caused by differences in the background subtraction applied to the two optical data sets. In this scenario the contribution of the stellar population, and thus the stellar mass of the cluster, may be higher (abridged).

The origin of the high X-ray luminosities of most ultraluminous X-ray sources (ULXs) still remains poorly understood. Most of the scenarios proposed to explain their nature assume that ULXs are powered by accretion on to a black hole (BH). In this case, the detection of compact radio emission and the location of the ULXs in the Fundamental Plane (X-ray versus radio-luminosity plane) can provide an estimate of the ULX BH mass and help address the question of their physical nature. We present the results of a high-resolution (very long baseline interferometry) radio observational campaign aimed at detecting and studying compact radio emission in four ULXs with known radio counterparts. We find that one of the targets (N4559-X4) was previously misclassified: its low X-ray luminosity indicates that the source is not a ULX. No milliarcsec-scale radio emission is detected for N4559-X4 nor for the ULXs N4490-X1 and N5194-X2, for which upper limits on the radio luminosities are estimated. These limits argue strongly against the presence of supermassive BHs in these three systems. For N4559-X4, the low X-ray luminosity and the ratio of the radio and X-ray luminosities suggest the presence of an X-ray binary. Compact radio emission is detected for the ULX N5457-X9 within its \textit{Chandra} positional error, making N5457-X9 a potential intermediate-mass BH with steady jet emission.

Ultraluminous X-ray sources (ULXs) were an enigma since their discovery. ASCA showed that they are accreting black holes (BHs) with unknown mass, while Chandra & XMM data shows that the many of the lower luminosity ULXs can be explained by extreme accretion onto stellar remnant BHs. The discovery of hyperluminous X-ray sources has led to specu- lation that we may, once again, have found the elusive intermediate mass black holes, yet in the last year, a third sub-class of extreme ULXs has emerged, with their nature, as yet, un- known. Here we define and discuss the sub-classes of these fascinating systems, exploring possible explanations for their nature, and indicating possible future steps in their analysis.

We present a multi-mission X-ray analysis of a bright (peak observed 0.3-10 keV luminosity of ~ 6x10^{40} erg s^{-1}), but relatively highly absorbed ULX in the edge-on spiral galaxy NGC 5907. The ULX is spectrally hard in X-rays (Gamma ~ 1.2-1.7, when fitted with an absorbed power-law), and has a previously-reported hard spectral break consistent with it being in the ultraluminous accretion state. It is also relatively highly absorbed for a ULX, with a column of ~ 0.4-0.9x10^{22} atom cm^{-2} in addition to the line-of-sight column in our Galaxy. Although its X-ray spectra are well represented by accretion disc models, its variability characteristics argue against this interpretation. The ULX spectra instead appear dominated by a cool, optically-thick Comptonising corona. We discuss how the measured 9 per cent rms variability and a hardening of the spectrum as its flux diminishes might be reconciled with the effects of a very massive, radiatively-driven wind, and subtle changes in the corona respectively. We speculate that the cool disc-like spectral component thought to be produced by the wind in other ULXs may be missing from the observed spectrum due to a combination of a low temperature (~ 0.1 keV), and the high column to the ULX. We find no evidence, other than its extreme X-ray luminosity, for the presence of an intermediate mass black hole (~ 10^2 - 10^4 Msun) in this object. Rather, the observations can be consistently explained by a massive (greater than ~ 20 Msun) stellar remnant black hole in a super-Eddington accretion state.

Ultracompact X-ray binaries (UCXBs) are thought to have relatively simple binary evolution post-contact, leading to clear predictions of their luminosity function. We test these predictions by studying the long-term behavior of known UCXBs in our Galaxy, principally using data from the MAXI All-Sky Survey and the Galactic Bulge Scans with RXTE's PCA instrument. Strong luminosity variations are common (and well-documented) among persistent UCXBs, which requires an explanation other than the disk instability mechanism. We measure the luminosity function of known UCXBs in the Milky Way, which extends to lower luminosities than some proposed theoretical luminosity functions of UCXBs. The difference between field and globular cluster X-ray luminosity functions in other galaxies cannot be explained by an increased fraction of UCXBs in globular clusters. Instead, our measured luminosity function suggests that UCXBs only make up a small fraction of the X-ray binaries above a few times 10^36 ergs/s in both old field populations and globular clusters.

We study the mass transfer rates and disk stability conditions of ultracompact X-ray binaries (UCXBs) using empirical time-averaged X-ray luminosities from Paper I (Cartwright et al. 2013) and compiled information from the literature. The majority of UCXBs are consistent with evolutionary tracks for white dwarf donors. Three UCXBs with orbital periods longer than 40 minutes have mass transfer rates above 10^{-10} Msun/year, inconsistent with white dwarf donor tracks. We show that if helium star donors can retain their initial high entropy, they can explain the observed mass transfer rates of these UCXBs. Several UCXBs show persistent luminosities apparently below the disk instability limit for irradiated He accretion disks. We point out that a predominantly C and/or O disk (as observed in the optical spectra of several) lowers the disk instability limit, explaining this disagreement. The orbital period and low time-averaged mass transfer rate of 2S 0918-549 provide evidence that the donor star is a low-entropy C/O white dwarf, consistent with optical spectra. We combine existing information to constrain the masses of the donors in 4U 1916-053 (0.064+-0.010 Msun) and 4U 1626-67 (<0.036 Msun for a 1.4 Msun neutron star). We show that 4U 1626-67 is indeed persistent, and not undergoing a transient outburst, leaving He star models as the best explanation for the donor.

We present a photometric survey of the optical counterparts of ultraluminous X-ray sources (ULXs) observed with the Hubble Space Telescope in nearby (< 5 Mpc) galaxies. Of the 33 ULXs with Hubble & Chandra data, 9 have no visible counterpart, placing limits on their M_V of ~ -4 to -9, enabling us to rule out O-type companions in 4 cases. The refined positions of two ULXs place them in the nucleus of their host galaxy. They are removed from our sample. Of the 22 remaining ULXs, 13 have one possible optical counterpart, while multiple are visible within the error regions of other ULXs. By calculating the number of chance coincidences, we estimate that 13 +/- 5 are the true counterparts. We attempt to constrain the nature of the companions by fitting the SED and M_V to obtain candidate spectral types. We can rule out O-type companions in 20 cases, while we find that one ULX (NGC 253 ULX2) excludes all OB-type companions. Fitting with X-ray irradiated models provides constraints on the donor star mass and radius. For 7 ULXs, we are able to impose inclination-dependent upper and/or lower limits on the black holes mass, if the extinction to the assumed companion star is not larger than the Galactic column. These are NGC 55 ULX1, NGC 253 ULX1, NGC 253 ULX2, NGC 253 XMM6, Ho IX X-1, IC342 X-1 & NGC 5204 X-1. This suggests that 10 ULXs do not have O companions, while none of the 18 fitted rule out B-type companions.

The high stellar densities in the cores of globular clusters cause significant stellar interactions. These stellar interactions can produce close binary mass-transferring systems involving compact objects and their progeny, such as X-ray binaries and radio millisecond pulsars. Comparing the numbers of these systems and interaction rates in different clusters drives our understanding of how cluster parameters affect the production of close binaries. In this paper we estimate stellar encounter rates (Gamma) for 124 Galactic globular clusters based on observational data as opposed to the methods previously employed, which assumed "King-model" profiles for all clusters. By deprojecting cluster surface brightness profiles to estimate luminosity density profiles, we treat "King-model" and "core-collapsed" clusters in the same way. In addition, we use Monte-Carlo simulations to investigate the effects of uncertainties in various observational parameters (distance, reddening, surface brightness) on Gamma, producing the first catalog of GC stellar encounter rates with estimated errors. Comparing our results with published observations of likely products of stellar interactions (numbers of X-ray binaries, numbers of radio millisecond pulsars, and gamma-ray luminosity) we find both clear correlations and some differences with published results.

We present the results from an X-ray and optical study of a new sample of eight extreme luminosity ultraluminous X-ray source (ULX) candidates, which were selected as the brightest ULXs (with L_X > 5x10^40 erg/s) located within 100 Mpc identified in a cross correlation of the 2XMM-DR1 and RC3 catalogues. These objects are so luminous that they are difficult to describe with current models of super-Eddington accretion onto all but the most massive stellar remnants; hence they are amongst the most plausible candidates to host larger, intermediate-mass black holes (IMBHs). Two objects are luminous enough in at least one observation to be classed as hyperluminous X-ray source (HLX) candidates, including one persistent HLX in an S0 galaxy that (at 3x10^41 erg/s) is the second most luminous HLX yet detected. The remaining seven sources are located in spiral galaxies, and several appear to be closely associated with regions of star formation as is common for many less luminous ULXs. However, the X-ray characteristics of these extreme ULXs appear to diverge from the less luminous objects. They are typically harder, possessing absorbed power-law continuum spectra with photon indexes ~ 1.7, and are potentially more variable on short timescales, with data consistent with ~ 10-20 per cent rms variability on timescales of 0.2-2 ks. These properties appear consistent with the sub-Eddington hard state, which given the observed luminosities of these objects suggests the presence of IMBHs with masses in the range 10^3-10^4 M_Sun. As such, this strengthens the case for these brightest ULXs as good candidates for the eventual conclusive detection of the highly elusive IMBHs. However, we caution that a combination of the highest plausible super-Eddington accretion rates and the largest permitted stellar black hole remnants cannot be ruled out without future, improved observations.

We studied the optical counterpart of the intermediate-mass black hole candidate HLX-1 in ESO 243-49. We used a set of Very Large Telescope imaging observations from 2010 November, integrated by Swift X-ray data from the same epoch. We measured standard Vega brightnesses U = 23.89 +/- 0.18 mag, B = 25.19 +/- 0.30 mag, V = 24.79 +/- 0.34 mag and R = 24.71 +/- 0.40 mag. Therefore, the source was ~1 mag fainter in each band than in a set of Hubble Space Telescope images taken a couple of months earlier, when the X-ray flux was a factor of 2 higher. We conclude that during the 2010 September observations, the optical counterpart was dominated by emission from an irradiated disk (which responds to the varying X-ray luminosity), rather than by a star cluster around the black hole (which would not change). We modelled the Comptonized, irradiated X-ray spectrum of the disk, and found that the optical luminosity and colours in the 2010 November data are still consistent with emission from the irradiated disk, with a characteristic outer radius r_{out} ~ 2800 r_{in} ~ 10^{13} cm and a reprocessing fraction ~ 2 x 10^{-3}. The optical colours are also consistent with a stellar population with age <~ 6 Myr (at solar metallicity) and mass ~ 10^4 M_{sun}; this is only an upper limit to the mass, if there is also a significant contribution from an irradiated disk. We strongly rule out the presence of a young super-star-cluster, which would be too bright. An old globular cluster might be associated with HLX-1, as long as its mass <~ 2 x 10^6 M_{sun} for an age of 10 Gyr, but it cannot significantly contribute to the observed very blue and variable optical/UV emission.

Ultraluminous X-ray sources (ULXs), first observed ~30 years ago, have been argued as extreme stellar mass black hole binaries or a new class intermediate mass black hole. In order to settle this debate, scientists have utilised a wide range of telescopes, exploiting large sections of the electro-magnetic spectrum. Here we review some of the insight gained from these observational studies, collating an overview of our current position in ULX research.

We explore the nature of ultraluminous X-ray sources through detailed investigations of their spectral shape using some of the highest quality data available in the XMM-Newton public archives. Phenomenological models allow us to characterise their spectra, while more 'physically-motivated' models enable us to explore the physical processes underlying these characteristics. These physical models imply the presence of extreme (probably super-Eddington) accretion on to stellar mass black holes.

We present the previously unanalysed high quality XMM-Newton spectrum of an ultraluminous X-ray source candidate in NGC 4517. As with other high quality ULX spectra, a downturn in the spectrum is observed at ~6 keV. Both of the recent disc reflection and Comptonisation interpretations of this feature are applied, in order to present a direct comparison, and are found to provide statistically equivalent representations of the current data. We find that the reflection model requires the accretion disc to have a highly super-solar iron abundance, while the Comptonisation model requires low temperature Comptonising electrons, and for the corona to be optically thick. These physical requirements are discussed in detail, and physically motivated scenarios are highlighted in which each model can be considered a viable explanation for the observed emission. By extending our consideration of these two interpretations to high energies, we demonstrate that observations of ULXs at energies above ~10 keV should be extremely useful when attempting to distinguish between them. With current instrumentation, it is only viable to perform these observations for M82 X-1, but future high angular resolution hard X-ray imaging spectrometers, such as the Hard X-ray Imaging System due to fly on Astro-H, should go a long way to resolving this issue.

We present the preliminary results of two Gemini campaigns to constrain the mass of the black hole in an ultraluminous X-ray source (ULX) via optical spectroscopy. Pilot studies of the optical counterparts of a number of ULXs revealed two candidates for further detailed study, based on the presence of a broad He II 4686 Angstrom emission line. A sequence of 10 long-slit spectra were obtained for each object, and the velocity shift of the ULX counterpart measured. Although radial velocity variations are observed, they are not sinusoidal, and no mass function is obtained. However, the broad He II line is highly variable on timescales shorter than a day. If associated with the reprocessing of X-rays in the accretion disc, its breadth implies that the disc must be close to face-on.

New optical narrowband imaging observations of the fields of several ULXs are presented. Known supershell nebulae are associated with a number of these ULXs, which we detect in emission line filters such as [S II], He II, [O II] and [O III]. New nebulae are discovered, which are candidate ULX-powered supershells. The morphologies and emission line fluxes of these nebulae could then be used to infer the properties of the emitting gas, which gives clues to the energizing source (photoionization and/or shock-excitation, both possibly from the ULX). Studies of supershells powered by ULXs can help to constrain the nature of ULXs themselves, such as the isotropy of the X-ray emission and the strength of their outflows.

Since their discovery, Ultraluminous X-ray sources (ULXs) have attracted attention due to their combination of extreme luminosities and extra-nuclear locations. However, they are a fairly rare phenomenon, and attempts to investigate the general properties of the population have been hindered by a relative lack of known sources. Here, we present a large catalogue of ULX candidates including 655 detections of 475 discrete sources, based on the 2XMM Serendipitous Survey. To demonstrate the potential of such a resource, we present some scientific analysis of this population, focusing on the spectral turnover seen, often at ~6 keV, in the highest quality ULX data. We also demonstrate how the recent reflection and Comptonisation interpretations of this feature may be distinguished observationally in the future, specifically using NGC 4517 ULX1, a previously unanalysed source with high quality data, as an example case.

(Abridged) We revisit the question of the nature of ULXs through a detailed investigation of their spectral shape, using the highest quality X-ray data available in the XMM-Newton public archives. We confirm that simple spectral models commonly used for the analysis and interpretation of ULXs (power-law continuum and multi-colour disc blackbody models) are inadequate in the face of such high quality data. Instead we find two near ubiquitous features in the spectrum: a soft excess and a roll-over in the spectrum at energies above 3keV. We investigate a range of more physical models to describe these data. We find that disc plus Comptonised corona models fit the data well, but the derived corona is cool, and optically thick (tau ~ 5-30). We argue that these observed disc temperatures are not a good indicator of the black hole mass as the powerful, optically thick corona drains energy from the inner disc, and obscures it. We estimate the intrinsic (corona-less) disc temperature, and demonstrate that in most cases it lies in the regime of stellar mass black holes. These objects have spectra which range from those similar to the highest mass accretion rate states in Galactic binaries, to those which clearly have two peaks, one at energies below 1 keV (from the outer, unComptonised disc) and one above 3 keV (from the Comptonised, inner disc). However, a few ULXs have a significantly cooler corrected disc temperature; we suggest that these are the most extreme stellar mass black hole accretors, in which a massive wind completely envelopes the inner disc regions, creating a cool photosphere. We conclude that ULXs provide us with an observational template for the transition between Eddington and super-Eddington accretion flows, with the latter occupying a new ultraluminous accretion state.

We report the results of spectral and temporal variability studies of the ultraluminous X-ray sources (ULXs) contained within the interacting pair of galaxies NGC 4485/4490, combining Chandra and XMM-Newton observations. Each of the four separate observations provide at least modest quality spectra and light curves for each of the six previously identified ULXs in this system; we also note the presence of a new transient ULX in the most recent observation. No short-term variability was observed for any ULX within our sample, but three out of five sources show correlated flux/spectral changes over longer timescales, with two others remaining stable in spectrum and luminosity over a period of at least five years. We model the spectra with simple power-law and multi-colour disc black body models. Although the data is insufficient to statistically distinguish models in each epoch, those better modelled (in terms of their chi^2 fit) by a multi-colour disc black body appear to show a disc-like correlation between luminosity and temperature, whereas those modelled by a power-law veer sharply away from such a relationship. The ULXs with possible correlated flux/spectral changes appear to change spectral form at ~2 x 10^39 erg s^{-1}, suggestive of a possible change in spectral state at high luminosities. If this transition is occurring between the very high state and a super-Eddington ultraluminous state, it indicates that the mass of the black holes in these ULXs is around 10-15 M_solar.

We systematically analyze all the available X-ray spectra of disc accreting neutron stars (atolls and millisecond pulsars) from the RXTE database. We show that while these all have similar spectral evolution as a function of mass accretion rate, there are also subtle differences. There are two different types of hard/soft transition, those where the spectrum softens at all energies, leading to a diagonal track on a colour-colour diagram, and those where only the higher energy spectrum softens, giving a vertical track. The luminosity at which the transition occurs is correlated with this spectral behaviour, with the vertical transition at L/L_Edd ~0.02 while the diagonal one is at ~0.1. Superimposed on this is the well known hysteresis effect, but we show that classic, large scale hysteresis occurs only in the outbursting sources, indicating that its origin is in the dramatic rate of change of mass accretion rate during the disc instability. We show that the long term mass accretion rate correlates with the transition behaviour, and speculate that this is due to the magnetic field being able to emerge from the neutron star surface for low average mass accretion rates. While this is not strong enough to collimate the flow except in the millisecond pulsars, its presence may affect the inner accretion flow by changing the properties of the jet.