By experimentally manipulating the time when young marsh tits, Parus palustris, could become established, I tested the influence of size, hatching date, social dominance in caged situations, and time of season on establishment success. Individuals that managed to become established were divided into two groups: those that became established in new territories and those that were found in the same territory as the one where they were caught. In no circumstances did size, hatching date or social dominance influence establishment success. The only factor tested that had a significant effect on establishment success, was when the juveniles were released; those that established themselves in new territories had been released earlier than unsuccessful ones. Thus, prior occupancy rather than hatching date per se is the factor determining establishment in new territories. Juveniles that became re-established in their former territory had been released significantly later than those not becoming re-established. As the season progresses, increasingly fewer unestablished juveniles are present to fill vacancies. Thus, late-released juveniles stood a better chance of finding their flock position still vacant compared with juveniles released early, whose previous positions would already have been occupied by other juveniles.