5. On the nature of the condensations at the tips of the Eagle Nebula
To summarise the main observational characteristics of the clumps at the tips of the fingers; three compact, massive, dense and cold clumps have been detected close to the tips of the fingers. These clumps have no detectable compact embedded near or mid-infrared sources, radio continuum counterparts or molecular outflows. The crucial question is whether these are clumps formed by the action of a radiatively driven implosion (RDI - Lefloch & Lazareff 1994, 1995) at the finger tips, or are instead pre-existing dense cores that are either Class 0 protostars, or clumps at an earlier stage of evolution. André (1996) has proposed several criteria to define Class 0 protostars: a) they should show indirect evidence of the presence of a central YSO - for example by the detection of a compact centimetre wavelength free-free emission peak or by the presence of a collimated outflow or nearby water maser clumps, b) they should have centrally peaked but extended submillimetre continuum emission tracing the presence of a spheroidal circumstellar dust envelope, c) they should exhibit a high ratio of submillimetre to bolometric luminosity: , where is the luminosity radiated longward of 350 µm (i.e. the spectral energy distribution should resemble a single temperature blackbody at T 15-20 K). For we estimate , and . It is clear however that a) does not match the characteristics of the Eagle clumps, and that our estimates of the pressure balance between the ionisation front and the cloud interior indicate that RDI has not yet led to the compression of the finger tips (see Sect. 6). It therefore seems inescapable that they are pre-existing clumps at a very early stage of evolution. In particular no protostellar core has yet developed within the cores which could be capable of producing the UV radiation which would be needed to generate free-free radio emission. Neither has there apparently been time for a molecular outflow to develop. The other characteristics also rule out `Class I'-`Class III' objects (see André 1996 for more detailed definitions of these categories).
Although there have been claims of detections of almost 30 `Class 0' protostars (André 1996), there are few, if any, convincing detections of `protostellar' objects earlier than `Class 0'. Other, similar objects such as the protostellar condensations in NGC2024 were originally claimed by Mezger et al. (1992) to be isothermally cold ( 20 K), massive protostars. However, subsequent IR and spectroscopic studies challenged this interpretation, following the detection of associated water maser and molecular outflow activity (summarised in Wiesemeyer et al. 1997). Perhaps the best candidates for such an early class of objects have been the so-called `prestellar' cores (Ward-Thompson et al. 1994, 1997). These `starless cores' are characterised by having relatively dense cores, no outflows, and a shallow density distribution. However, it is far from clear that they are collapsing; in particular the amount of magnetic support within their cores is unknown, and the few available determinations of core temperatures are highly dependent of how much of the gas has become frozen onto, or depleted onto grains (Henriksen et al. 1997). We therefore believe that the condensations at the tips of the Eagle Nebula are amongst the best candidates so far reported as being in the earliest stage of Class 0 protostellar development.
© European Southern Observatory (ESO) 1999
Online publication: December 22, 1998