Abstract. We present new spectroscopic and photometric observations of the young Galactic open cluster Westerlund 1 (Wd 1) that reveal a unique population of massive evolved stars. We identify ∼200 cluster members and present spectroscopic classifications for ∼25% of these. We find that all stars so classified are unambiguously post-Main Sequence objects, consistent with an apparent lack of an identifiable Main Sequence in our photometric data to V ∼ 20. We are able to identify rich populations of Wolf Rayet stars, OB supergiants and short lived transitional objects. Of these, the latter group consists of both hot (Luminous Blue Variable and extreme B supergiant) and cool (Yellow Hypergiant and Red Supergiant) objects -we find that half the known Galactic population of YHGs resides within Wd 1. We obtain a mean V − M V ∼ 25 mag from the cluster Yellow Hypergiants, implying a Main Sequence turnoff at or below M V = −5 (O7 V or later). Based solely on the masses inferred for the 53 spectroscopically classified stars, we determine an absolute minimum mass of ∼1.5 × 10 3 M for Wd 1. However, considering the complete photometrically and spectroscopically selected cluster population and adopting a Kroupa IMF we infer a likely mass for Wd 1 of ∼10 5 M , noting that inevitable source confusion and incompleteness are likely to render this an underestimate. As such, Wd 1 is the most massive compact young cluster yet identified in the Local Group, with a mass exceeding that of Galactic Centre clusters such as the Arches and Quintuplet. Indeed, the luminosity, inferred mass and compact nature of Wd 1 are comparable with those of Super Star Clusters -previously identified only in external galaxies -and is consistent with expectations for a Globular Cluster progenitor.

Aims. We aim to characterise the properties of a third massive, red supergiant dominated galactic cluster. Methods. To accomplish this we utilised a combination of near/mid-IR photometry and spectroscopy to identify and classify the properties of cluster members, and statistical arguments to determine the mass of the cluster. Results. We found a total of 16 strong candidates for cluster membership, for which formal classification of a subset yields spectral types from K3-M4 Ia and luminosities between log(L/L ) ∼ 4.5-4.8 for an adopted distance of 6 ± 1 kpc. For an age in the range of 16-20 Myr, the implied mass is 2-4×10 4 M , making it one of the most massive young clusters in the Galaxy. This discovery supports the hypothesis that a significant burst of star formation occurred at the base of Scutum-Crux arm between 10-20 Myr ago, yielding a stellar complex comprising at least ∼10 5 M of stars (noting that since the cluster identification criteria rely on the presence of RSGs, we suspect that the true stellar yield will be significantly higher). We highlight the apparent absence of X-ray binaries within the star formation complex and finally, given the physical association of at least two pulsars with this region, discuss the implications of this finding for stellar evolution and the production and properties of neutron stars.

Abstract. We utilise Midcourse Space Experiment mid-IR imaging and published data to discuss the (massive) star formation region at galactic longitude ∼305 • , apparently associated with the Wolf Rayet WR 48a and the attendant clusters Danks 1 and 2. A spectacular three lobed wind blown bubble surrounds the aforementioned sources, for which we may infer a minimum age of ∼3 Myr from the presence of the WCL star. Near IR data reveals the presence of numerous embedded sources on the periphery of the wind blown bubble. The presence of coincindent H 2 O, OH and methanol maser emission is suggestive of ongoing massive star formation, which is suppported by the fluxes of the associated IR sources, and the requisite LyC flux required to support the emission from the subset that have associated ucH  regions. Consideration of the integrated radio flux of the complex implies that a minimum of 31 O7V stars must be present, under the assumption of no photon leakage. Given the age and morphology of the complex and in particular the observation that the central exciting clusters have entirely cleared their natal material, we expect this assumption will be violated, and hence that the true population of massive stars is likely to be significantly larger. If confirmed, the G305 complex represents one of the most massive regions of ongoing triggered star formation currently identified in the galaxy.

Aims. Multiwavelength observations of the young massive cluster Westerlund 1 have revealed evidence for a large number of OB supergiant and Wolf-Rayet binaries. However, in most cases these findings are based on the detection of secondary binary characteristics, such as hard X-ray emission and/or non-thermal radio spectra and hence provide little information on binary properties such as mass ratio and orbital period. To overcome this shortcoming we have initiated a long temporal baseline, multi-epoch radial velocity survey that will provide the first direct constraints on these parameters. Methods. VLT/FLAMES+GIRAFFE observations of Wd1 were made on seven epochs from late-June to early-September 2008, covering ∼35 confirmed members of Wd1 and ∼70 photometrically-selected candidate members. Each target was observed on a minimum of three epochs, with brighter cluster members observed on five (or, in a few cases, seven) occasions. Individual spectra cover the 8484-9001 Å range, and strong Paschen-series absorption lines are used to measure radial velocity changes in order to identify candidate binary systems for follow-up study.Results. This study presents first-epoch results from twenty of the most luminous supergiant stars in Wd1. Four new OB supergiant members of Wd1 are identified, while statistically significant radial velocity changes are detected in ∼60% of the targets. W43a is identified as a short-period binary, while W234 and the newly-identified cluster member W3003 are probable binaries and W2a is a strong binary candidate. The cool hypergiants W243 and W265 display photospheric pulsations, while a number of early-mid B supergiants display significant radial velocity changes of ∼15-25 km s −1 that we cannot distinguish between orbital or photospheric motion in our initial short-baseline survey. When combined with existing observations, we find 30% of our sample to be binary (6/20) while additional candidate binaries support a binary fraction amongst Wd1 supergiants in excess of ∼40%, a figure that is likely to increase as further data become available.

The G305 H ii complex (G305.4+0.1) is one of the most massive star‐forming structures yet identified within the Galaxy. It is host to many massive stars at all stages of formation and evolution, from embedded molecular cores to post‐main‐sequence stars. Here, we present a detailed near‐infrared analysis of the two central star clusters Danks 1 and Danks 2, using Hubble Space Telescope+NICMOS imaging and Very Large Telescope+ISAAC spectroscopy. We find that the spectrophotometric distance to the clusters is consistent with the kinematic distance to the G305 complex, an average of all measurements giving a distance of 3.8 ± 0.6 kpc. From analysis of the stellar populations and the pre‐main‐sequence stars, we find that Danks 2 is the elder of the two clusters, with an age of 3+3− 1 Myr. Danks 1 is clearly younger with an age of 1.5+1.5− 0.5 Myr, and is dominated by three very luminous H‐rich Wolf–Rayet stars which may have masses ≳100 M⊙. The two clusters have mass functions consistent with the Salpeter slope, and total cluster masses of 8000 ± 1500 and 3000 ± 800 M⊙ for Danks 1 and Danks 2, respectively. Danks 1 is significantly the more compact cluster of the two, and is one of the densest clusters in the Galaxy with log (ρ/M⊙ pc−3) = 5.5+0.5− 0.4. In addition to the clusters, there is a population of apparently isolated Wolf–Rayet stars within the molecular cloud’s cavity. Our results suggest that the star‐forming history of G305 began with the formation of Danks 2, and subsequently Danks 1, with the origin of the diffuse evolved population currently uncertain. Together, the massive stars at the centre of the G305 region appear to be clearing away what is left of the natal cloud, triggering a further generation of star formation at the cloud’s periphery.

We present Very Large Telescope/FORS2 time‐series spectroscopy of the Wolf–Rayet (WR) star #41 in the Sculptor group galaxy NGC 300. We confirm a physical association with NGC 300 X‐1, since radial velocity variations of the He iiλ4686 line indicate an orbital period of 32.3 ± 0.2 h which agrees at the 2σ level with the X‐ray period from Carpano et al. We measure a radial velocity semi‐amplitude of 267 ± 8 km s−1, from which a mass function of 2.6 ± 0.3 M⊙ is obtained. A revised spectroscopic mass for the WN‐type companion of 26+7−5 M⊙ yields a black hole mass of 20 ± 4 M⊙ for a preferred inclination of 60°–75°. If the WR star provides half of the measured visual continuum flux, a reduced WR (black hole) mass of 15+4−2.5 M⊙ (14.5+3−2.5 M⊙) would be inferred. As such, #41/NGC 300 X‐1 represents only the second extragalactic WR plus black hole binary system, after IC 10 X‐1. In addition, the compact object responsible for NGC 300 X‐1 is the second highest stellar‐mass black hole known to date, exceeded only by IC 10 X‐1.

Abstract. We present a compilation of spectroscopic observations of the sgB[e] star CI Cam, the optical counterpart of XTE J0421+560. This includes data from before, during, and after its 1998 outburst, with quantitative results spanning 37 years. The object shows a rich emission line spectrum originating from circumstellar material, rendering it difficult to determine the nature of either star involved or the cause of the outburst. We collate all available pre-outburst data to determine the state of the system before this occurred and provide a baseline for comparison with outburst and post-outburst data. During the outburst all lines become stronger, and hydrogen and helium lines become significantly broader and asymmetric. After the outburst, spectral changes persist for at least three years, with Fe  and [N ] lines still a factor of ∼2 above the pre-outburst level and He , He , and N  lines suppressed by a factor of 2-10. We find that the spectral properties of CI Cam are similar to other sgB[e] stars and therefore suggest that the geometry of the circumstellar material is similar to that proposed for the other objects: a two component outflow, with a fast, hot, rarefied polar wind indistinguishable from that of a normal supergiant and a dense, cooler equatorial outflow with a much lower velocity. Based on a comparison of the properties of CI Cam with the other sgB[e] stars we suggest that CI Cam is among the hotter members of the class and is viewed nearly pole-on. The nature of the compact object and the mechanism for the outburst remain uncertain, although it is likely that the compact object is a black hole or neutron star, and that the outburst was precipitated by its passage through the equatorial material. We suggest that this prompted a burst of supercritical accretion resulting in ejection of much of the material, which was later seen as an expanding radio remnant. The enhanced outburst emission most likely originated either directly from this supercritical accretion, or from the interaction of the expanding remnant with the equatorial material, or from a combination of both mechanisms.

We study the long‐term behaviour of the equatorial disc of the Be/X‐ray binary X Persei (X Per), combining new low‐resolution IUE spectra and IR photometry with UV, optical and IR observations collected from the literature. We find that the near‐UV continuum level of X Per varies along with the optical brightness. From the UV observations, we also find that during optical high states the flux excess due to the intrinsic stellar variability and/or electron scattering in the disc is at most 15–20 per cent of the photospheric flux. From the data taken in discless and near‐discless states (optical low states) we show that the stellar photosphere can be modelled with Teff=31000 K and log g=4. With this model we derive E(B−V)=0.39, and estimate the distance to X Per as 950±200 pc (assuming R*=9 R⊙). We fit the (quasi‐)simultaneous optical and IR photometry with a simple disc model including free–bound and free–free radiation. We find that the density of the disc at the photosphere of the star varies along with the brightness of X Per, and that in optical high states the disc in X Per is among the densest of all Be stars: ρ0=(1.5±0.3)×10−10 g cm−3. The disc density at the photosphere varies by a factor of at least 20 from optical high to low states. During disc build‐up and break‐down phases, and also in phases when the disc is relatively stable, we find a very steep radial density gradient of the disc of X Per. This may reflect the limitations of some of the assumptions in our model. We find that in a disc‐loss event the disc loses mass at a rate of about 5×10−9 M⊙ yr−1. For a disc build‐up phase we find a disc‐mass growth rate of about 4×10−9 M⊙ yr−1. This growth rate is consistent with a model that feeds the disc from the ‘ordinary’ mass‐loss of the star, but we cannot exclude the possibility that other phenomena contribute to the disc growth as well.