This article focuses on problems with the definition and empirical identification of immigrant “first” and “second” generations in the United States. These loosely conceived aggregates are decomposed into a typology of distinct generational cohorts defined by age and life stage at migration for the foreign born and by parental nativity for the U.S. born. Differences in educational and occupational attainment and in language and other aspects of acculturation are then examined to consider whether the practice of “lumping” these generational cohorts together, or “splitting” them into distinctive units of analysis, is empirically supported by available evidence. The paper concludes with some thoughts on data needs and methodological considerations in the study of immigrant generations.