Date Published: December 14, 2009
Publisher: Public Library of Science
Author(s): Craig D. H. Sherman, Erik Wapstra, Mats Olsson, Stewart Plaistow. http://doi.org/10.1371/journal.pone.0008252
Abstract: A large number of studies in postcopulatory sexual selection use paternity success as a proxy for fertilization success. However, selective mortality during embryonic development can lead to skews in paternity in situations of polyandry and sperm competition. Thus, when assessment of paternity fails to incorporate mortality skews during early ontogeny, this may interfere with correct interpretation of results and subsequent evolutionary inference. In a previous series of in vitro sperm competition experiments with amphibians (Litoria peronii), we showed skewed paternity patterns towards males more genetically similar to the female.
Partial Text: Patterns of paternity in offspring are routinely used to infer fertilization success for analysis of mating system evolution, genetic compatibility of mates, or innate competitive ability in sperm competition [e.g. 1], , . However, such studies are fraught with potential biases introduced by developmental arrest, selective abortion, or other forms of compromised ontogeny. Few studies have directly tested paternity at fertilization and followed it through various stages of ontogeny [although see 4], . This is particularly problematic in internally fertilizing species where it has been difficult to directly observe developing embryos. Externally fertilizing species are increasing being used in studies of sexual selection as they offer excellent systems for separating out genetic and nongenetic effects on fertilization success and offspring fitness as their external mode of fertilization makes them particularly amenable to controlled laboratory studies using in vitro fertilization [e.g.], , , , , [9,10]. These systems also allow for the direct assessment of fertilization success and mortality can be followed through various stages of ontogeny, something that is much less tractable in internally fertilizing species.
Despite recent cautioning on the use of paternity data to infer fertilization success in situations of sperm competition, our results show robust and conclusive support for the hypothesis that biased paternity success sampled late in the embryonic development of the Peron’s tree frog is consistent with the corresponding between-male difference in siring success at fertilization. While some clutches can show significant post-fertilization mortality, this does not appear to result from incompatibility effects between maternal and paternal genomes. Thus, in this species there appears to be no mortality skews taking effect in a progressive manner through ontogeny as would be expected if parental genetic incompatibility affects offspring probability of survival during that period of development, as demonstrated in mice  and some insects [e.g., 15].