Determinants of fitness in species with complex life cycles

Amphibians hold a unique position as the only vertebrate group where basically all members feature complex live cycles with two distinct free-living stages that have to face entirely different challenges in their respective environment. In the proposed project we want to investigate the determinants of fitness in the Neotropical frog Allobates femoralis, by exploring the effects of parental relatedness and parental investment on individual reproductive success at various life cycle stages and the associated patterns of space use. To this end we will install an experimental population of A. femoralis on a river island of ~5 ha in size. The island is situated in ~200 m distance to a population on the mainland where we have been conducting research from 2008 - 2011. Additionally, we will perform specific breeding and behavioural experiments with an ex-situ breeding population at the University of Vienna. The combined approach of controlled and natural field studies, and laboratory experiments will allow us to address questions on the effects of parental relatedness on fertilization success, larval survival in the clutch and survival of post-metamorphic juveniles. At the same time the island population and experiments in captivity allow us to answer questions on patterns of juvenile dispersal and adult spatial behaviour, as well as the use of spatial, visual, and acoustic information for parental behaviour and space use in males and females. The pan-Amazonian aromobatid frog Allobates femoralis is a highly suitable species for genetic studies at the individual and on the population level. Throughout the prolonged breeding period males are highly territorial and announce territory occupancy by their prominent advertisement call. Females are similarly iteroparous and show strong site fidelity. Courtship and mating take place inside a male`s territory. Tadpole transport to bodies of water is generally performed by A. femoralis males, with occasional cases of maternal transport. Analyses of parentage over two successive generations of adult individuals have identified a highly polygynandrous mating system, where males and females contribute in equal share to the progeny in the subsequent generation. A recent study did not find evidence for inbreeding avoidance or depression in A. femoralis but indications for the prevalence of an optimal genetic divergence between mating partners. What makes this proposed project unique is the opportunity to study a wild population of anurans in their natural environment during a controlled introduction on a river island. Island populations and controlled introductions are a proven setup in the investigation of mammal, bird and lizard populations, but have never been applied to amphibian species so far. The availability of a well studied reference population on the mainland as a backup and control, and the captive breeding population at the University of Vienna alleviate potential risks that come with this demanding approach.

In the project Determinants of fitness in species with complex life cycles we have investigated how sexual and natural selection shape reproductive behaviours in the Neotropical frog Allobates femoralis. To this end we have installed both, an ex-situ breeding population at the University of Vienna and an experimental in-situ population of A. femoralis on a river island in French Guiana, where we undertook population-wide genetic sampling as well as specific behavioural experiments under controlled conditions. The controlled introduction of 1800 genotyped tadpoles allowed us to evaluate the use of genetic fingerprinting for markrecapture studies across metamorphosis in amphibians. Our results show that microsatellites markers are a highly powerful tool for studying amphibian populations on an individual basis. The ability to track individual tadpoles throughout metamorphosis until adulthood will be of substantial value for future studies on amphibian population- and behavioural ecology and useful for species conservation projects. By molecular parentage analysis of tadpoles we were able to investigate the logistics of tadpole transport and the associated patterns of space use in this species. We found that A. femoralis males distribute tadpoles across several water bodies as a bet-hedging strategy; and we discovered that females flexibly compensate for missing male care. Experiments in the lab further revealed that males and females use different offspring discrimination strategies that may lead to either parental or aggressive (cannibalistic) behaviour. These findings demonstrate that poison frogs are highly flexible and capable of strategic planning when it comes to parental decision-making. The combination of spatial translocations, individual tracking, cross-foster experiments, and the use of the closed island setup further allowed us to investigate orientation abilities and homing performance of A. femoralis in both familiar and novel environments. Our findings suggest that poison frogs use spatial learning to improve orientation in their local area, and may also use olfaction to explore novel reproductive resources. We also undertook a supplementation experiment with artificial water bodies, which showed that suitable aquatic sites for tadpole deposition are a limited resource. After the installation of the pools, the population size almost doubled. These findings are of importance for the conservation of amphibians, as they highlight the role of non-trophic interactions between species and suggest intervention routes to mitigate the global amphibian decline.