Crested Tit Parus cristatus young from first broods dispersed 1 week later if they were born in small isolated pine plots (‘habitat fragments’) compared with individuals in a large pine forest (‘continuous habitat’). This delay in dispersal was caused by an extended period between fledging and dispersal. In second broods, the delay was even longer due to the interbrood interval being 9 days longer in habitat fragments. As nestlings in habitat fragments had a lower body-mass, and age at dispersal was negatively correlated with nestling body-mass within each nest, the postponed dispersal from fragments might be explained partly by a lower body-mass. Alternatively, postponed dispersal from fragments could result from a barrier effect caused by reluctance to cross inhospitable habitat.

Immigration by young from first broods into habitat fragments was delayed by approximately 3 weeks, and proportionally more second brood emigrants were recovered in this type of habitat. These results are in agreement with the hypothesis that fragments are second-choice habitat. Early immigrants into continuous habitat had a higher probability of settlement in winter flocks compared with late ones, independent of condition or age. Therefore, Crested Tits born in habitat fragments probably have a lower chance of settling in first-choice habitat.