The main focus of this thesis is the investigation of an interesting new population of massive passively evolving galaxies found at high redshifts. We use a sample of these galaxies at redshifts 1 <= z <= 2 drawn from the Gemini Deep Deep Survey sample and measure their structural parameters based on the Hubble Space Telescope imaging in the rest frame visible and ultraviolet wavelength range. Our analysis shows that a fraction of these systems are very compact, with effective radii of R_e < 1 kpc, even when observed in rest frame visible light. The average size of these objects is 2-5 times smaller than the typical size of an early-type galaxy of similar mass found locally.
We combine the results from our morphological analysis with data from published spectroscopic samples of quiescent systems with known structural parameters. Analysis of these data for galaxies over the redshift range 0 <= z <= 2.5 shows that passively evolving galaxies are continuously and gradually growing in size. We also find smooth evolution of the stellar mass density within the central kiloparsec of these systems. The stellar mass density grows by a factor of 3 from z = 0 to z = 2.5. None of the models proposed to drive the structural evolution of early-type objects can explain all the observed aspects of this process.
Because these massive compact galaxies have such small angular sizes, future studies of these systems will benefit from adaptive optics. In order to compile a large statistical sample of these objects suitable for adaptive optics follow-up, we first need to find a large number of targets with nearby bright natural guide stars. As a first step in this process, we describe the properties of a set of one square degree regions of the sky we have located that have a rare combination of high stellar surface density and low levels of extinction. We demonstrate that the adaptive optics-related properties of these fields are in some cases orders of magnitude better than those of existing deep fields.