2010ApJ...717L.103N

We present deep Keck spectroscopy for 17 morphologically selected field spheroidals in the redshift range 1.05 < z < 1.60 in order to investigate the continuity in physical properties between the claimed massive compact red galaxies ("nuggets") at z =~ 2 and well-established data for massive spheroidal galaxies below z =~ 1. By combining Keck-based stellar velocity dispersions with Hubble Space Telescope-based sizes, we find that the most massive systems (M_dyn>10¹¹ M_☉) grew in size over 0 < z < 1.6 as (1 + z)^{–0.75±0.10} (i.e., x2 since z = 1.5) whereas intermediate mass systems (10¹¹ M_☉>M_dyn>10¹⁰ M_☉) did not grow significantly. These trends are consistent with a picture in which more massive spheroidals formed at higher redshift via "wetter" mergers involving greater dissipation. To examine growth under the favored "dry" merger hypothesis, we also examine size growth at a fixed velocity dispersion. This test, uniquely possible with our dynamical data, allows us to consider the effects of "progenitor bias." Above our completeness limit (σ>200 km/s), we find size growth consistent with that inferred for the mass-selected sample, thus ruling out strong progenitor bias. To maintain continuity in the growth of massive galaxies over the past 10 Gyr, our new results imply that size evolution over 1.3 < z < 2.3, a period of 1.9 Gyr, must have been even more dramatic than hitherto claimed if the red sources at z>2 are truly massive and compact.