2011ApJ...726...97L

If star formation proceeds by thermal fragmentation and the subsequent gravitational collapse of the individual fragments, how is it possible to form fragments massive enough for O and B stars in a typical star-forming molecular cloud where the Jeans mass is about 1 M_☉ at the typical densities (10⁴/cm3) and temperatures (10 K)? We test the hypothesis that a first generation of low-mass stars may heat the gas enough that subsequent thermal fragmentation results in fragments ≥ 10 M_☉, sufficient to form B stars. We combine ATCA and Submillimeter Array observations of the massive star-forming region G8.68-0.37 with radiative transfer modeling to derive the present-day conditions in the region and use this to infer the conditions in the past, at the time of core formation. Assuming that the current mass/separation of the observed cores equals the fragmentation Jeans mass/length and the region's average density has not changed requires the gas temperature to have been 100 K at the time of fragmentation. The postulated first generation of low-mass stars would still be around today, but the number required to heat the cloud exceeds the limits imposed by the observations. Several lines of evidence suggest the observed cores in the region should eventually form O stars yet none have sufficient raw material. Even if feedback may have suppressed fragmentation, it was not sufficient to halt it to this extent. To develop into O stars, the cores must obtain additional mass from outside their observationally defined boundaries. The observations suggest that they are currently fed via infall from the very massive reservoir (∼1500 M_☉) of gas in the larger parsec scale cloud around the star-forming cores. This suggests that massive stars do not form in the collapse of individual massive fragments, but rather in smaller fragments that themselves continue to gain mass by accretion from larger scales.