New surveys with the Spitzer Space Telescope identify distant star-forming and active galaxies by their strong emission at far-infrared wavelengths, which provides strong constraints on these galaxies' bolometric energy. Using early results from Spitzer surveys at 24 μm, we argue that the faint sources correspond to the existence of a population of infrared-luminous galaxies at z ≳ 1 that are not expected from predictions based on previous observations from ISO and IRAS. Combining Spitzer images with deep ground-based optical and Hubble Space Telescope imaging, we discuss the properties of galaxies selected at 24 μm in the region of the Chandra Deep Field South, including redshift and morphological distributions. Galaxies with z ≲ 1 constitute roughly half of the faint 24 μm sources. Infrared-luminous galaxies at these redshifts span a wide variety of normal to strongly interacting/merging morphologies, which suggests that a range of mechanisms produce infrared activity. Large-area, joint surveys between Spitzer and HST are needed to understand the complex relation between galaxy morphology, structure, environment, and activity level, and how this evolves with cosmic time. We briefly discuss strategies for constructing surveys to maximize the legacy of these missions.