Properties of millimeter galaxies:: Constraints from K-band blank fields

We have used the IRAM Plateau de Bure millimeter interferometer to locate with subarcsecond accuracy the dust emission of three of the brightest 1.2 mm sources in the NTT Deep Field selected from our 1.2 mm Max-Planck-Millimeter Bolometer Array survey at the IRAM 30 m telescope. We combine these results with deep B to K imaging and Very Large Array interferometry. Reliable identifications are an essential step toward an understanding of the high-redshift (sub) millimeter galaxy population, toward testing the common belief that they are scaled-up analogs of local dusty ultraluminous galaxies, and in shedding light on the possible connection to spheroid formation. Strikingly, none of the three accurately located millimeter galaxies MM J120546-0741.5, MM J120539-0745.4, and MM J120517-0743.1 has a K-band counterpart down to the faint limit of K-s > 21.9. This implies that these three galaxies either are extremely obscured and/or are at very high redshifts (z greater than or similar to 4). We combine our results with literature data for 11 more (sub) millimeter galaxies that are identified with similar reliability. In terms of their K-band properties, the sample divides into three roughly equal groups: (1) undetected to K similar to 22, (2) detected in the near-infrared but not the optical, and (3) detected in the optical with the possibility of optical follow-up spectroscopy. We find a trend in this sample between near-infrared to submillimeter and submillimeter to radio spectral indexes, which in comparison to spectral energy distributions (SEDs) of low-redshift infrared-luminous galaxies suggests that the most plausible primary factor causing the extreme near-infrared faintness of our objects is their high redshift. We show that the near-infrared to radio SEDs of the sample are inconsistent with SEDs that resemble local far-infrared cool galaxies with moderate luminosities, which were proposed in some models of the submillimeter sky. We briefly discuss the implications of the results for our understanding of galaxy formation.