Introduction: Seed dispersal by animals is key for forest regeneration and passive restoration of tropical forests because it maintains plant diversity and accelerates community turnover. We propose that the assessment of seed dispersal by animals should be included in the management of secondary forests because it provides evidence of the reestablishment of species interactions and ecosystem functioning.

Objectives & predictions: In this study we examined shifts in the importance of different seed dispersal modes during regeneration by modeling the proportion of trees dispersed by bats, small birds, large birds, non-volant mammals and abiotic means as a function of forest age. Moreover, we predicted that the likelihood of a single plant species being dispersed by multiple dispersal modes (i.e., redundancy) increased over time as non-volant mammals and large birds become more important during regeneration.

Methods: We used data from saplings and trees collected on a chronosequence that spans >100 years of regeneration in the Barro Colorado Nature Monument. We assigned one or more dispersal modes to each species according to a large data set collected in the last decades at the same site. Moreover, we generated a dispersal redundancy index where higher values indicated that a greater proportion of modes dispersed each species.

Results: Contrary to expectations, tree species dispersed by non-volant mammals dominated from early stages of regeneration, and recovered to old growth levels after 40 years post-abandonment. Seed dispersal by small birds declined over time during regeneration, while bat dispersal played a relatively minor role in structuring the tree community. Redundancy of dispersal modes across species of saplings and trees significantly increased with forest age, but the change in magnitude was small.

Implications/Conclusions: Our results suggest that proximity to old growth forests coupled with low hunting, explained the quick recovery of seed dispersal by animals in the Barro Colorado Nature Monument, and especially dispersal by non-volant mammals. We suggest that aspects of seed dispersal should be monitored when managing secondary forests (e.g., while employing an Adaptive Management Cycle).