The Magellanic Clouds, consisting of the LMC, the SMC and the Bridge offer an ideal laboratory for studying cluster formation in a lower metal abundance environment at an unrivaled closeness to us among external galaxies. It is known that very young populous clusters like R136 are still being formed in the LMC, where populous clusters include more than 10000 stars tightly gravitationally bound. In this talk, I will present mm and sub-mm CO observations of the Magellanic molecular clouds obtained with NANTEN, SEST, and ASTE at spatial resolutions of 5–50pc. I will then use these CO data to identify the parent cloud cores for populous clusters and discuss the cluster formation by comparing the cloud properties with those of the Milky Way.

Stars form by gravoturbulent fragmentation of interstellar gas clouds. The supersonic turbulence ubiquitously observed in Galactic molecular gas generates strong density fluctuations with gravity taking over in the densest and most massive regions. Collapse sets in to build up stars and star clusters.

Turbulence plays a dual role. On global scales it provides support, while at the same time it can promote local collapse. Stellar birth is thus intimately linked to the dynamic behavior of parental gas clouds, which governs when and where protostellar cores form, and how they contract and grow in mass via accretion from the surrounding cloud material to build up stars. The equation of state plays a pivotal role in the fragmentation process. Under typical cloud conditions, massive stars form as part of dense clusters following the “normal” mass function observed, e.g. in the solar neighborhood. However, for gas with an effective polytropic index greater than unity star formation becomes biased towards isolated massive stars. This is relevant for understanding the properties of zero-metallicity stars (Population III) or stars that form under extreme environmental conditions like in the Galactic center or in luminous starbursts.

Various physical processes are believed to trigger star formation on the borders of Galactic HII regions. Among these, the collect & collapse process is particularly attractive as it allows the formation of massive objects (single stars or clusters). In order to identify specific cases of this way of triggering star formation we are carrying out a multi-wavelength study of Galactic HII regions that exhibit signposts of massive-star formation at their borders. Hereby, we present two typical examples of such sources and discuss the results in the framework of the collect and collapse process, which seems to be at work as the major triggering agent in these two cases.

We are conducting a survey of several regions of high-mass star formation to assess their content and structure. The observations include spitzer observations, ground-based optical and near-IR imaging surveys, and optical and IR spectra of objects and locations in the molecular clouds. The goal of the survey is to gain a better understanding of the processes involved in high mass star formation by determining the characteristics of the stars detected in these regions and investigating the properties of the interstellar medium (ISM) environment in which these stars form. In this contribution, we present results on the identification and spatial analysis of young stars in three clusters, W5/AFGL 4029, S255, and S235. First we show how the IRAC data are used to roughly segregate young stars according to their mid-infrared colors, into two groups corresponding the SED class I and class II young stellar objects. Then using the IRAC data in combination with 2MASS, we show how more young stars can be identified. Finally, we examine the spatial distributions of young stars in these clusters and find a range of morphologies and of peak surface densities.

The HST Treasury Program on the Orion Nebula Cluster has been recently completed (May 2005). Using 104 orbits of HST time we have imaged a field ${\sim }1/6$ of a square degree nearly centered on the Trapezium stars. The survey, made with ACS, WFPC2 and NICMOS-Camera 3 in parallel, has imaged this cornerstone region with unprecedented sensitivity (23-24 mag), dynamic range (${\sim}12$ mag), spatial resolution (50mas), and wide spectral coverage (9 filters from U to H). We have assembled the richest, most accurate and unbiased dataset of stellar photometry for pre-main-sequence objects ever obtained, an essential tool for understanding of the star formation process in regions dominated by massive OB stars.

About 20% of the high mass O-type stars in the Galaxy are found outside stellar clusters and associations. In the solar neighbourhood this fraction amounts to 43 O-type stars. In the framework of high-mass star formation, we search for the origins of these stars. We aim at separating the O-type field stars from runaway O-type stars, where the former does not betray any indication of past dynamical interactions resulting in their present field location. We specifically search for the presence of stellar clusters, but also evaluate spatial velocities, distance from Galactic plane, locations in the Galaxy with respect to young stellar clusters. Among our sample of 43, we find that 5 stars are in fact observed in the K-band to be members of previously unnoticed small clusters. On the other hand it is plausible that nearly half the sample (22 objects) could have undergone a dynamical ejection from a young cluster. Based on the current available data, ${\sim }4%$ of the Galactic O-type stars cannot be associated with a stellar cluster. Finally by assuming that stars form in clusters with a power law in membership number with index −1.7 down to “clusters” consisting of single stars, we calculate the expected statistics regarding isolated O stars and O stars in OB-associations. We conclude that the results of the calculations are consistent with the observed statistics of O-type stars.

We describe a programme that aims to increase the known sample of massive young stellar objects (MYSOs) by an order of magnitude. About 2000 candidates colour-selected from the MSX survey are being followed up at radio, mm and IR wavelengths to identify genuine MYSOs from the UCHII regions and other contaminants. Results so far indicate that the strategy does indeed deliver a significant fraction of luminous YSOs that will provide the basis for future galaxy-wide systematic studies.

We have recently obtained moderate resolution (R${\sim}$8,000–12,000) high signal-to-noise H- and K-band spectroscopy of a number of optically visible, well studied OB stars (Hanson et al. 2005) to test the reliability of a pure near-infrared quantitative analysis (Repolust et al. 2005). The analysis of 25 of these OB stars via near-infrared spectra alone using the NLTE line-blanketed model atmosphere code FASTWIND (Puls et al. 2005) has proved successful at constraining stellar and wind parameters, consistent with results from previous optical analyses of these stars. This opens the door to quantitative analysis of OB stars at an extraordinarily young age, while they are still heavily enshrouded in their birth cocoons. Because the analysis requires good quality spectra at both H and K band, present 8-m class telescopes limit us to sources which are not extremely embedded ($A_V < 30$). As a first example, we present a preliminary analysis of the heavily reddened ($A_V = 25$), early-O star ionizing the UCHII region, G29.96-0.02. Challenges facing such an analysis include contributions from excess thermal emission from circumstellar material (disks, etc.) which weaken or even eliminate photospheric lines used in the analysis, nebular contamination in several of the principle H and He lines and crowding or general confusion in these very young and typically complex regions. Spectrographs coupled with state of the art adaptive optics will be extremely useful in minimizing these challenges, and may allow even fairly complex regions to be directly studied.

We present initial results from a survey of the Orion A and B molecular clouds made with the InfraRed Array Camera (IRAC) onboard the Spitzer Space Telescope. This survey encompasses a total of 5.6 square degrees with the sensitivity to detect objects below the hydrogen burning limit at an age of 1 Myr. These observations cover a number of known star forming regions, from the massive star forming clusters in the Orion Nebula and NGC 2024, to small groups of low mass stars in the L1641. We combine the IRAC photometry with photometry from the 2MASS point source catalog and use the resulting seven band data to identify stars with infrared excesses due to dusty disks and envelopes. Using the presence of an infrared excess as an indicator of youth, we show the distribution of young stars and protostars in the two molecular clouds. We find that roughly half of the stars are found in dense clusters surrounding the two regions of recent massive star formation in the Orion clouds, NGC 2024 and the Orion Nebula.

This review discusses near- and mid- infrared observations of Ultracompact (UC) HII regions. The importance of ISO mid-IR fine-structure nebular lines is emphasised, since only a small fraction of UCHII regions are observed directly in the near-IR. The reliability of contemporary atmospheric models for such indirect diagnostics is discussed, whilst a revised spectral type-temperature calibration is presented for Galactic O3 to B3 dwarfs. In particular, fine-structure line derived properties of G29.96–0.02 differ from the direct near-IR spectroscopic result and represents a serious discrepancy which needs to be addressed. Mid-IR MSX and Spitzer imaging permits the identification of those UCHII regions for which far-IR IRAS fluxes are reliable, relevant to the single versus cluster nature of individual sources. High spatial resolution imaging with ground-based 8m telescopes allows more direct tests, as recently applied to G70.29+1.60. Finally, recent Spitzer mid-IR observations of Giant HII regions are briefly discussed.

The Spitzer Space Telescope is very efficient at moderate-depth infrared mapping. The Galactic Legacy Infrared Midplane Survey Extraordinaire (GLIMPSE) and other projects have now imaged most of the massive star formation regions in the inner Galaxy. In this case, quantity is quality–the large datasets allow statistical and global analysis of star forming regions that was previously difficult. Data such as the GLIMPSE survey is proving useful to study different evolutionary stages of massive star formation, from pre-protostellar clouds through feedback and triggered star formation. We discuss selected team results relevant to each of these stages: mapping infrared dark clouds and the mid-IR extinction law, studies using radiative transfer models of the entire protostellar population of star forming regions such as M16 and M17 (see also Whitney, this conference), feedback and star formation in the G305 and RCW79 giant HII regions, and GLIMPSE's comment on even older objects such as debris disks and PN. Continuing similar analysis will hopefully lead to better understanding of large-scale questions of Galactic star formation such as the total star formation rate and the relationships between star formation and Galactic structure.

We present results from our ongoing survey of Galactic Giant HII Regions in the near-infrared. The luminosity function indicates that the IMF is uniform, independent of the galactocentric distance and compatible with Salpeter (1955) slope. Distances measured by the spectroscopic parallax method are systematically smaller than kinematic distances derived from radio techniques. As a consequence, the number of ionizing photons and the star formation rate is much lower than that derived from rotation models. Although the luminosities of the giant HII regions obtained from this method are lower than by other methods, the morphological type of the Milky Way is well in line of previous results, close to Sbc or Sc types.

Super star clusters are one of the most extreme star forming environments in the universe, and the most massive and dense of these may be proto globular clusters. Like individual massive stars, the earliest stages of super star cluster evolution are deeply obscured, and therefore our knowledge about their birth environments is currently very incomplete. However, the study of natal super star clusters has become somewhat of a cottage industry in recent years, and the sample of such objects has been growing rapidly with high-quality long-wavelength data now available from a number of observatories. The natal super star clusters identified in thermal-infrared and radio observations represent the youngest stage of massive star cluster evolution yet observed. Their properties appear to be similar to those of ultracompact hII regions in the milky way, but scaled up in total mass and luminosity. I will overview what we think we know about these objects based on existing observations, discuss their relationship to ultracompact hII regions, present new models of their spectral energy distributions based on 3-d simulations, and outline some of the most significant gaps in our current understanding.

Theoretical considerations lead to the expectation that stars should not have masses larger than about $m_{\rm max*}=60$–$120M_\odot$, while the observational evidence has been ambiguous. Only very recently has a physical stellar mass limit near $150M_\odot$ emerged thanks to modern high-resolution observations of local star-burst clusters. But this limit does not appear to depend on metallicity, in contradiction to theory. Important uncertainties remain though. It is now also emerging that star-clusters limit the masses of their constituent stars, such that a well-defined relation between the mass of the most massive star in a cluster and the cluster mass, $m_{\rm max}={\cal F}(M_{\rm ecl}) \le m_{\rm max*}\approx 150M_\odot$, exists. One rather startling finding is that the observational data strongly favour clusters being built-up by consecutively forming more-massive stars until the most massive stars terminate further star-formation. The relation also implies that composite populations, which consist of many star clusters, most of which may be dissolved, must have steeper composite IMFs than simple stellar populations such as are found in individual clusters. Thus, for example, $10^5$ Taurus–Auriga star-forming groups, each with 20 stars, will ever only sample the IMF below about $1M_\odot$. This IMF will therefore not be identical to the IMF of one cluster with $2\times 10^6$ stars. The implication is that the star-formation history of a galaxy critically determines its integrated galaxial IMF and thus the total number of supernovae per star and its chemical enrichment history. Galaxy formation and evolution models that rely on an invariant IMF would be wrong.

We discuss the expected properties of the first stellar generations in the Universe. We find that it is possible to discern truly primordial populations from the next generation of stars by measuring the metallicity of high-z star forming objects through the detection and the study of the emission line spectrum from HII regions associated to primordial massive stars. Moreover, the intensity of HeII emission lines can be used as a diagnostic to reveal the presence of massive stars and determine their importance relative to lower mass stars. Thanks to its very low background, the future James Webb Space Telescope (JWST) will be able to image and study first-light sources at very high redshifts. On the other hand, JWST's relatively small collecting area limits its capability in obtaining spectra of z$\,{\sim}\,$10–15 first-light sources to either the bright end of their luminosity function or to strongly lensed sources. With a suitable investment of observing time JWST will be able to detect individual Population III supernovae, thus identifying the very first stars that formed in the Universe.

I describe some common themes that emerged from the meeting, ending with some thoughts on connecting our studies with those of other galaxies.

I begin with a very brief summary of the nature of the physics problems facing us in massive star birth, and present a couple of (now belated) admonitions to the participants. I will say a few words about several of the high spots from papers on the newly born massive stars that caught my attention. This is thus a very personal and highly biased summary. There is no way I could mention all of the really new and interesting results presented at this meeting and I apologize in advance for leaving out many fine reports.