Context. Hatchell et al. (2005, A&A, 440, 151) (Paper I) published a submillimetre continuum map of the Perseus molecular cloud, detecting the starless and protostellar cores within it. Aims. To determine the evolutionary stage of each submm core in Perseus, and investigate the lifetimes of these phases. Methods. We compile spectral energy distributions (SEDs) from 2MASS (1-2 µm), Spitzer IRAC (3.6, 4.5, 5.8, 8.0 µm), Michelle (11 and 18 µm), IRAS (12, 25, 60, 100 µm), SCUBA (450 and 850 µm) and Bolocam (1100 µm) data. Sources are classified starless/protostellar on the basis of infrared and/or outflow detections and Class I/Class 0 on the basis of T bol , L bol /L smm and F 3.6 /F 850 . In order to investigate the dependence of these evolutionary indicators on mass, we construct radiative transfer models of Class 0 sources. Results. Of the submm cores, 56/103 (54%) are confirmed protostars on the basis of infrared emission or molecular outflows. Of these, 22 are classified Class 1 on the basis of three evolutionary indicators, 34 are Class 0, and the remaining 47 are assumed starless. Perseus contains a much greater fraction of Class 0 sources than either Taurus or Rho Oph. We derive estimates for the correlation between bolometric luminosity and envelope mass for Class I and Class 0 sources. Conclusions. Comparing the protostellar with the T Tauri population, the lifetime of the protostellar phase in Perseus is 0.25−0.67 Myr (95% confidence limits). The relative lifetime of the Class 0 and Class 1 phases are similar, confirming the results of Visser et al. (2002, AJ, 124, 2756 in isolated cores. We find that for the same source geometry but different masses, evolutionary indicators such as T bol vary their value. It is therefore not always appropriate to use a fixed threshold to separate Class 0 and Class I sources. More modelling is required to determine the observational characteristics of the Class 0/Class I boundary over a range of masses.

Aims. Solving the continuum radiative transfer equation in high opacity media requires sophisticated numerical tools. In order to test the reliability of such tools, we present a benchmark of radiative transfer codes in a 2D disc configuration. Methods. We test the accuracy of seven independently developed radiative transfer codes by comparing the temperature structures, spectral energy distributions, scattered light images, and linear polarisation maps that each model predicts for a variety of disc opacities and viewing angles. The test cases have been chosen to be numerically challenging, with midplane optical depths up 10 6 , a sharp density transition at the inner edge and complex scattering matrices. We also review recent progress in the implementation of the Monte Carlo method that allow an efficient solution to these kinds of problems and discuss the advantages and limitations of Monte Carlo codes compared to those of discrete ordinate codes. Results. For each of the test cases, the predicted results from the radiative transfer codes are within good agreement. The results indicate that these codes can be confidently used to interpret present and future observations of protoplanetary discs.

We have conducted a programme to determine the fundamental parameters of a substantial number of eclipsing binaries of spectral types O and B in the Small Magellanic Cloud (SMC). New spectroscopic data, obtained with the two‐degree‐field (2dF) multi‐object spectrograph on the 3.9‐m Anglo‐Australian Telescope, have been used in conjunction with photometry from the Optical Gravitational Lens Experiment (OGLE‐II) data base of SMC eclipsing binaries. Previously we reported results for 10 systems; in this second and concluding paper we present spectral types, masses, radii, temperatures, surface gravities and luminosities for the components of a further 40 binaries. The uncertainties are typically ±10 per cent on masses, ±4 per cent on radii and ±0.07 on log L. The full sample of 50 OB‐type eclipsing systems is the largest single set of fundamental parameters determined for high‐mass binaries in any galaxy. We find that 21 of the systems studied are in detached configurations, 28 are in semidetached post‐mass‐transfer states, and one is a contact binary. The overall properties of the detached systems are consistent with theoretical models for the evolution of single stars with SMC metal abundances (Z≃ 0.004); in particular, observed and evolutionary masses are in excellent agreement. Although there are no directly applicable published models, the overall properties of the semidetached systems are consistent with them being in the slow phase of mass transfer in case A. About 40 per cent of these semidetached systems show photometric evidence of orbital‐phase‐dependent absorption by a gas stream falling from the inner Lagrangian point on the secondary star towards the primary star. This sample demonstrates that case‐A mass transfer is a common occurrence amongst high‐mass binaries with initial orbital periods P≲ 5 d, and that this slow phase has a comparable duration to the detached phase preceding it. Each system provides a primary distance indicator. We find a mean distance modulus to the SMC of 18.91 ± 0.03 ± 0.1 (internal and external uncertainties; D= 60.6 ± 1.0 ± 2.8 kpc). This value represents one of the most precise available determinations of the distance to the SMC.

We report the detection, through spectropolarimetric observations, of a strong dipolar magnetic field of presumably fossil origin at the surface of the very young O star θ1 Ori C. The Stokes V signatures we detect are variable with time, the variations being consistent with rotational modulation. A detailed modelling of our observations indicates that this dipole field has an intensity of 1.1±0.1 kG and is inclined at 42°±6° with respect to the rotation axis (assumed to be inclined at 45° to the line of sight). We find, in particular, that the positive magnetic pole comes closest to the observer when the variable Hα emission component observed on this star reaches maximum strength. This discovery represents the first definite detection of a magnetic field in an O star, as well as the first detection of a fossil field in a very young star. We also investigate in this paper the magnetic confinement of the radiatively driven wind of θ1 Ori C in the context of the magnetically confined wind‐shock model of Babel & Montmerle. In the case of θ1 Ori C, this model predicts the formation of a large magnetosphere (extending as far as 2–3R∗), consisting of a very hot post‐shock region (with temperatures in excess of 10 MK and densities of about 1010–1011 cm‐3) generated by the strong collision of the wind streams from both stellar magnetic hemispheres, as well as a dense cooling disc forming in the magnetospheric equator. We find that this model includes most of the physics required to obtain a satisfactory level of agreement with the extensive data sets available for θ1 Ori C in the literature (and, in particular, with the recent X‐ray data and the phase‐resolved spectroscopic observations of ultraviolet and optical wind lines) provided that the mass‐loss rate of θ1 Ori C is at least 5 times smaller than that predicted by radiatively driven wind models. We finally show how new observations with the XMM or Chandra spacecraft could help us constrain this model much more tightly and thus obtain a clear picture of how magnetic fields can influence the winds of hot stars.

From observations collected with the ESPaDOnS spectropolarimeter, we report the discovery of magnetic fields at the surface of the mildly accreting classical T Tauri star (cTTS) V2129 Oph. Zeeman signatures are detected, both in photospheric lines and in the emission lines formed at the base of the accretion funnels linking the disc to the protostar, and monitored over the whole rotation cycle of V2129 Oph. We observe that rotational modulation dominates the temporal variations of both unpolarized and circularly polarized line profiles. We reconstruct the large‐scale magnetic topology at the surface of V2129 Oph from both sets of Zeeman signatures simultaneously. We find it to be rather complex, with a dominant octupolar component and a weak dipole of strengths 1.2 and 0.35 kG, respectively, both slightly tilted with respect to the rotation axis. The large‐scale field is anchored in a pair of 2‐kG unipolar radial field spots located at high latitudes and coinciding with cool dark polar spots at photospheric level. This large‐scale field geometry is unusually complex compared to those of non‐accreting cool active subgiants with moderate rotation rates. As an illustration, we provide a first attempt at modelling the magnetospheric topology and accretion funnels of V2129 Oph using field extrapolation. We find that the magnetosphere of V2129 Oph must extend to about 7R★ to ensure that the footpoints of accretion funnels coincide with the high‐latitude accretion spots on the stellar surface. It suggests that the stellar magnetic field succeeds in coupling to the accretion disc as far out as the corotation radius, and could possibly explain the slow rotation of V2129 Oph. The magnetospheric geometry we derive qualitatively reproduces the modulation of Balmer lines and produces X‐ray coronal fluxes typical of those observed in cTTSs.