A new hypothesis for the importance of seed dispersal in time

Adriana Guzmán,Pablo R Stevenson

doi:10.15517/rbt.v59i4.33186

Abstract

Most studies on seed dispersal in time have focused on seed dormancy and the physiological triggers for germination. However, seed dispersed by animals with low metabolic and moving rates, and long gut-passage times such as terrestrial turtles, could be considered another type of dispersal in time. This study tests the hypothesis that seeds dispersed in time may lower predation rates. We predicted that seeds deposited below parent trees after fruiting fall has finished is advantageous to minimize seed predators and should show higher survival rates. Four Amazonian plant species, Dicranostyles ampla, Oenocarpus bataua, Guatteria atabapensis and Ocotea floribunda, were tested for seed survival probabilities in two periods: during fruiting and 10-21 days after fruiting. Experiments were carried out in two biological stations located in the Colombian Amazon (Caparú and Zafire Biological Stations). Seed predation was high and mainly caused by non-vertebrates. Out of the four plant species tested, only Guatteria atabapensis supported the time escape hypothesis. For this species, seed predation by vertebrates after the fruiting period increased (from 4.1% to 9.2%) while seed predation by non-vertebrates decreased (from 54.0% to 40.2%). In contrast, seed predation by vertebrates and by non-vertebrates after the fruiting period in D. ampla increased (from 7.9% to 22.8% and from 40.4% to 50.6%, respectively), suggesting predator satiation. Results suggest that for some species dispersal in time could be advantageous to avoid some type of seed predators. Escape in time could be an additional dimension in which seeds may reach adequate sites for recruitment. Thus, future studies should be address to better understand the survival advantages given by an endozoochory time-dispersal process.

Highlights

The Janzen-Connell hypothesis proposes that seeds have higher survival probabilities when they escape distance-and density-dependent predation below parental trees, affecting plant recruitment and diversity (Janzen 1970, Connell 1971)
This study addresses the following questions: (1) Are seed removal rates below parent trees lower after the fruiting period has finished than during fruiting period? (2) Do different seed predators have the same mortality effect at different phenophases? (3) If there are differences in mortality rates during and after fruiting, can they be associated to invertebrate or mammalian seed predation? In this study we experimentally test the hypothesis that seeds could minimize predation in time and according to this idea we predicted that seed predation will be higher during the fruiting period than after the fruiting period, affecting seed survival
Non-vertebrate seed predation of D. ampla was mainly caused by leaf cutter ants (Atta sp.) while seeds of O. floribunda were repeatedly buried by a medium sized beetle (0.5-1.0cm long), making it difficult to distinguish if seeds were predated or manipulated by invertebrates

Summary

Introduction

The Janzen-Connell hypothesis proposes that seeds have higher survival probabilities when they escape distance-and density-dependent predation below parental trees, affecting plant recruitment and diversity (Janzen 1970, Connell 1971). Mean retention time for Chelonoidis denticulata was estimated by Guzmán & Stevenson (2008) to be of 21 days, and for C. denticualta and C. carbonaria by Strong (2005) to range between 10 to 28 days with a peak at 11-15 days These observations might suggest that reptiles, especially highly frugivorous chelonians, could play an important role in the dispersal of seeds in time. Vertebrate and invertebrate seed predators influence distance- and density-dependent seed mortality (Wright 2002) These predators may act in a distance- or density-responsive manner, depending on the season, plant species or resource availability among other factors, with varying effects on seed mortality (Schupp 2007). A possible explanation for this is that invertebrates or pathogens often complete their relatively short life cycles within the area occupied by a large tree, while mammals do it over much larger areas in the course of an annual cycle (Terborgh & Wright 1994, Pizo 1997, Zipparro & Morellato 2005)

Methods

Results

Conclusion