Conceptual bases of intervention and policy are often subject to paradoxical dilemmas, such as the need for a coherent program model versus recognition of the diversity of a target population. This article aims to identify underlying paradoxical bases for expansion and replication of early childhood care and education programs, and to suggest potential resolutions of these paradoxes. Clarification of such paradoxes may provide guidance for the many decision points which arise in processes of identifying, expanding, and replicating early childhood programs based on evidence of quality or success. First, a brief history of early childhood care and education in the U.S. from the viewpoint of expansion and replication is presented. A five-fold typology of expansion and replication processes is proposed: staged replication, franchised replication, multi-site demonstrations, mandated replication, and government-supported private sector expansion. Second, paradoxes associated with the process of replication are considered, such as: (1) the ''can it work'' versus ''does it work'' paradox; (2) the fidelity versus local cultural relevance paradox; (3) the replication versus addition paradox; (4) the replication versus program improvement paradox; and (5) the representativeness versus feasibility paradox. In a concluding section, recommendations for funders, policy makers, and evaluators are made regarding next steps in expansion and replication of early childhood care and education programs.