Does primate disappearnace influcence the advanced regeneration assemblage in rainforest fragments

Background/Question/Methods

Primates represent important seed dispersers, especially of large-seeded tree species; however, information on the impact these animals have on advanced regeneration in forest fragments remains obscure. We tested the hypothesis that forest regeneration is limited in fragments without primates. During 17-months we recorded species richness and density of all juvenile plants already established as advanced regeneration (woody plants 5–100 cm tall and < 1 cm in basal diameter) in 3 forest fragments occupied (OF) by primates, 3 fragments unoccupied (UF) by primates, and 3 areas of continuous forest (CF) in the Lacandona rainforest, Mexico. Plant species were classified as vertebrate-dispersed (VD), primate-dispersed (PD), and abiotically-dispersed (AD). We assessed the influence of adult assemblages, primate presence, fragment attributes (size, age, distance to continuous forest), and canopy openness on advanced regeneration assemblages.

Results/Conclusions

We registered 6,879 advanced regeneration plants (2087 in CF, 1472 in OF, and 3320 in UF) belonging to 133 species, 59 genera, 37 families in 405 1-m² plots. Sixty-nine % were VD species, 20% were AD, and 11% were PD. Assemblage structure of adult trees and advanced regeneration plants differed among habitat types, depending on the seed dispersal mode. Abundance of advanced regeneration PD species showed a similar pattern to that found for adult trees. Density of adult trees and advanced regeneration plants in each dispersal mode differed between habitat types. Richness of adult trees and advanced regeneration plants in each dispersal mode also differed between habitat types. Considering all dispersal modes, richness of adult trees was higher in UF and CF than in OF, while richness of advanced regeneration plants was higher in UF than in CF and OF. Richness of PD adult trees was higher in CF than in OF and UF, but richness of PD advanced regeneration plants was similar among habitat types. Proportion of adult trees in each dispersal mode corresponded to the proportion of advanced regeneration plants found in each habitat type. Multiple correlation analysis showed the density of PD plants was highly correlated with adult tree density and to a lesser degree with habitat size (positively) and distance to continuous forest (negatively). None of the ecological variables measured influenced the richness of PD plants. We conclude that primate disappearance constrains the recruitment of large-seeded species, because advanced regeneration assemblages were largely explained by adult tree composition. Nevertheless, we cannot reject the possibility that particular PD species may be affected by the disappearance of primates.