The Evolutionary Ecology and Sexual Selection of a
Madagascan Poison Frog (Mantella laevigata)
Dissertation by Heather E. Heying
Abstract:
Mantella laevigata is a toxic Madagascan frog (Mantellinae: Ranidae: Anura) that is aposematically colored, small, terrestrial, diurnal, and that breeds in rain-filled phytotelmata. In the field, I undertook the first study of its social system and evolutionary ecology, and the ecology of members of the treehole guild with whom it shares oviposition sites. I reviewed the literature on reproductive modes and parental care in anurans, noting that tropical, terrestrial-breeding species (of which phytotelm-breeders are a subset) have evolved parental care more often than other species.
During 925 hours of focal watches, I discerned several complex behaviors, including male territoriality; extended, stereotyped courtships; and maternal feeding of trophic eggs to tadpoles. Clutch size is one. Larval crane-flies (tipulids) predate Mantella eggs, and two other treehole-breeding anurans (Plethodontohyla notostica and Anodonthyla boulengeri, both microhylids) compete for space with Mantella. Convergence with the neotropical dendrobatids is discussed.
Experiments revealed that oviposition sites are limiting for Mantella. I discussed the evolution of phytotelm-breeding in light of this result, as well as risks including desiccation, predation, cannibalism and competition. I surveyed 53 wells over several months and found that Mantella oviposition is positively correlated with low pH; short, wet wells; and few tipulids, conspecific tadpoles, or other species of frogs. P. notostica adults protect preexisting Mantella eggs, but discourage new oviposition. After Mantella metamorphs disperse from a well, reproductive activity increases. I delineated a foodweb of the phytotelmata of northeastern Madagascar.
The theoretical core of this research is sexual selection. Experimentally, females were given choices between calls of individual males ("good genes"), of varying length calls ("good current condition"), in different quality territories ("good resources"). My previous observations of females scouting for oviposition sites before courtships, males leaving courtships to engage in territorial defense of oviposition sites, and biparental care, suggested the "good resources hypothesis." This hypothesis was supported, and I discuss why this system fits a model more common to birds than anurans. Finally, I assessed mechanisms of male reproductive success, which included three distinct male territorial strategies, prior-resident advantage in predicting the outcome of fights, paternal care, and manipulation of male-female conflict.
Ph.D. in Biology earned from the University of Michigan, 2001.
Members of my doctoral committee:
Arnold Kluge (chair)
Dick Alexander
Ron Nussbaum
Barbara Smuts
