Fuzzy relationships among plant dispersal mechanisms, syndromes, and animal vectors

Knowledge of which animal clades and dispersal mechanisms contribute to seed dispersal is critical for understanding the ecology and evolution of plant communities. However, a broad understanding is limited by a focus on single clades and mechanisms. Here, we analyzed a Europe-wide multi-clade seed dispersal dataset with interactions among 447 animal species and 1901 plant species via four dispersal mechanisms: endozoochory (ingestion followed by egestion), epizoochory (external attachment), synzoochory (actively carried by vertebrates), and myrmecochory (dispersal by ants). We aimed to (1) characterize the relative importance of animal dispersal clades and biotic dispersal mechanisms; (2) evaluate links among clades and mechanisms; (3) compare dispersal of dry- and fleshy-fruited diaspores; and (4) evaluate the predictive power of seed dispersal syndromes with empirical data on observed dispersal mechanisms. This revealed that endozoochory by birds and mammals encompasses 90% of plant-animal interaction records. Mammal species tend to have more plants depending upon their dispersal services; however, due to the larger diversity of bird species, more plants depend upon the cumulative services of birds (followed by mammals, insects, and reptiles). Plants with fleshy fruits comprise only 15% of all plant species dispersed by animals but had more animal partners per plant species than plants with dry fruits. Fleshy-fruited plant species were more specifically selected by dispersers, while selection of dry-fruited plants depended more on their relative abundance. Different animal clades tend to disperse plant species via different mechanisms. Endozoochory is the most recorded interaction for birds, mammals, and reptiles. Epizoochory is almost exclusively carried out by mammals, with a negligible role of birds, insects, and reptiles. Synzoochory is performed chiefly by ants (via myrmecochory), followed by birds and then mammals. Predicting vector importance by assigning dispersal syndromes based on diaspore traits (including endozoochorous, epizoochorous, synzoochorous, and myrmecochorous syndromes) only moderately predicted empirically observed dispersal mechanisms. Animals disperse many more species than inferred from diaspore traits, including plants with abiotic syndromes and plants without clearly visible morphological traits. Our results suggest that different animal clades perform complementary dispersal services via multiple mechanisms to plant species with fleshy as well as dry fruits.