Previous
abstract 
Next
abstract
We will present moderate resolution ($\delta\lambda$/ $\lambda \approx$ 600), high signal-to-noise 2 micron spectra of 26 stars in the region of M17. The purpose of our $K$-band classification system was to investigate hot stars embedded in shrouded HII regions and study formation processes and early evolution of the most massive stars. We will show that such spectra are excellent probes of the stellar population in HII regions behind large extinction. In our first survey of the M17 region we have found five morphologically different spectra: 1) spectra which show atomic photospheric lines of normal hot stars (5 new O-type stars have been identified so far); 2) spectra which show no features through out the entire $K$-band; 3) spectra showing CO overtone emission (and sometimes B$\gamma$ emission) 4) a pure nebular spectrum, due to a compact HII region and 5) spectra showing strong CO overtone absorption features, due to interloping cool field giants. Combining the recently obtained spectral information with the $JHK$ colors and images and $V$-band images of the field, has allowed us to develop an evolutionary scenario for the M17 region.
A large number of the brightest stellar sources in M17 show excess 2 micron emission relative to their $J$ and $H$ magnitudes when compared to normal reddened stellar atmospheres. These where the stars that appeared either completely featureless or showed pronounced CO bandheads in emission. These objects may be enshrouded in a dense pre-main sequence disk or shell, which is being picked up in the $K$-band as excess emission and completely obscuring the underlying star. In some cases the disk or shell is warm and dense enough ($\rho \approx$ 10$^{10}$ cm$^{-3}$) to excite the CO overtone bandheads into emission. Our initial survey was not deep enough to search for an underlying B-star population that we expect to be present given a normal initial mass function and the 5 O-type stars already found. A lack of lower mass normal main sequence stars could explain the very luminous pre-main sequence stars we have found.
