Primates and Dung Beetles: Two Dispersers Are Better than One in Secondary Forest

Abstract

Primary seed dispersal by primates (phase I) followed by secondary seed dispersal by dung beetles (phase II) is a common diplochorous system in tropical forests. In such systems, phase I affects the occurrence/outcome of phase II, triggering cascading effects along the chain of plant recruitment with direct consequences on seed dispersal effectiveness. However, we know very little regarding whether seed dispersal effectiveness is increased or decreased by phase II and whether this effect is consistent among habitats. Using a primate–dung beetle diplochorous system, we determined 1) the characteristics of phase I that may affect phase II; 2) the pathways relating biotic/abiotic factors to seed/seedling survival; and 3) if the direction and/or magnitude of phase II effects on seed dispersal effectiveness depend on phase I characteristics. We marked and characterized the dispersal characteristics of 981 seeds dispersed by two tamarin species (Saguinus mystax, Leontocebus nigrifrons) and checked the fate of 503 of them for ≥1 year. Seeds dispersed by L. nigrifrons and seeds surrounded by larger amounts of dung were more likely to be buried by dung beetles. Burial increased seed survival in secondary forest while low seed density increased germination in both habitats. Seed burial increased seed dispersal effectiveness more strongly in secondary (+52.2%) vs. in primary forest (+5.0%), in L. nigrifrons (+12.9%) vs. in S. mystax (+7.9%) feces, and in larger fecal portions (+22.1%) vs. in small–medium ones (+7.3–7.4%). In conclusion, two seed dispersers are more effective than one only in secondary forest, and the magnitude of increase of seed dispersal effectiveness with phase II depends on how the seeds are primarily dispersed.