ABSTRACT. Massive stars are important in a variety of contexts: individually they are examples of stars that undergo significant mass loss which is driven by line radiation pressure; as denizens of the Galaxy they play a significant role in the galactic recycling effort; as ensembles of stars they contain clues to massive star evolution in general and the question of supernova progenitors in particular. Finally, under extreme conditions of pathological star formation massive stars participate in the star burst phenomena seen in some galaxies. If we wish to understand star bursts, we must first understand locally observed variations in the mass spectrum. But all of these diverse areas of study rely on certain fundamental parameters of the stars themselves-here we include temperature, luminosity, mass, and chemical composition. And central to accurate determinations of these parameters are the distances to the stars. This review will focus on what is known about the loose clustering of massive stars termed OB associations. These associations contain many, but by no means all, of the stars with masses greater than 15Mq. Most of our knowledge of the stellar content of the OB associations in our own Galaxy is surprisingly sketchy due to their large angular size, but large-field CCD detectors are changing this. Magellanic Cloud associations can now be compared with OB associations in our own Galaxy, and the initial mass function shows surprising variations. The physical cause of these variations is still unknown. Another development is the study of embedded star clusters: infrared array detectors have been probing the buried stars in molecular clouds and finding an unexpectedly large population of these objects. Both these young IR clusters, and a sprinkling of evolved red supergiants, suggest that the star-formation history of OB associations may be more complicated than we have supposed.