This study aimed to identify potential factors responsible for geographically structured morphological variation within the widespread Australian frogs Limnodynastes tasmaniensis Günther and L. peronii Duméril & Bibron. There was support for James's rule, and both latitude and present climate explained large amounts of the variation in body size and shape (particularly in L. peronii ). There was also some support for the influence of several biogeographical barriers. Finally, both species were sexually dimorphic for body size and the degree of sexual size dimorphism (SSD) varied geographically. Climate was an important explanation for SSD variation in L. peronii , while latitude was most important for L. tasmaniensis . Geographical variations in sexual selection via male-male physical competition and climate-related resources are suggested as potential explanations for SSD variation in L. peronii .

Marine turtle hatchlings emerge from nests on oceanic beaches, crawl to the surf zone and migrate offshore. Predators in shallow water can take many hatchlings, but once the turtles reach deeper water, both encounters with predators and mortality rates probably decline. Behavioral studies have demonstrated that hatchlings show changes in swimming activity as they migrate offshore. During the first 24 h of migration, most species swim continuously (the 'frenzy'), thereby minimizing their time in shallow waters; however, swimming activity later declines in duration and vigor, especially at night (the 'postfrenzy'). One interpretation of these differences is that hatchling migratory behavior evolved in response to the threat of predators (the 'predation risk' hypothesis). To further test this hypothesis, we quantified the daily swimming activity shown by the flatback Natator depressus, the only marine turtle that lacks an oceanic phase in its development. Instead, the hatchlings remain in relatively shallow Australian (continental shelf) waters where they may frequently encounter predators. We speculated that because of these encounters, flatback hatchlings might have evolved activity patterns that show little, if any, decline during migration. Over 4 d of laboratory observations, flatback activity at night declined by <13%. Over the same time period, nocturnal activity declined by 60 to 95% (depending on species) in other marine turtles with an oceanic phase in their life history. Our data therefore support the hypothesis that predation played an important role in shaping the evolution of hatchling migratory behavior, although they do not provide direct evidence of a cause and effect relationship.

We investigated the phylogeography of two closely related Australian frog species from open forest habitats, Limnodynastes tasmaniensis and L. pemnii, using mitochondria1 ND4 sequence data. Comparison of our results with previous work on Litoria fallax allowed us to test the generality of phylogeographic patterns among nonrainforest anurans along the east coast of Australia. In general, there was no strong evidence for congruence between overall patterns of genetic structure in the three species. However, phylogenetic breaks congruent with the position of the Burdekin Gap were detected at some level in all species. As previously noted for closed forest taxa, this area of dry habitat appears to have been an important influence on the evolution of several open forest taxa. There were broad geographic similarities in the phylogenetic structuring of southern populations of L. pemnii and L. tasmaniensis. Contrarily, although the McPherson Range has previously been noted to coincide geographically with a major mtDNA phylogenetic break in Litoria fallax this pattern is not apparent in L. pemnii or L. tasmaniensis. It appears that major phylogeographic splits within L. pemnii and L. tasmaniensis may predate the Quaternary. We conclude that phylogeographies of open forest frogs are complex and more difficult to predict than for rainforest taxa, mainly due to an absence of palaeomodels for historical distributions of non-rainforest habitats.

No abstract

We investigated the phylogeography of two closely related Australian frog species from open forest habitats, Limnodynastes tasmaniensis and L. pemnii, using mitochondria1 ND4 sequence data. Comparison of our results with previous work on Litoria fallax allowed us to test the generality of phylogeographic patterns among nonrainforest anurans along the east coast of Australia. In general, there was no strong evidence for congruence between overall patterns of genetic structure in the three species. However, phylogenetic breaks congruent with the position of the Burdekin Gap were detected at some level in all species. As previously noted for closed forest taxa, this area of dry habitat appears to have been an important influence on the evolution of several open forest taxa. There were broad geographic similarities in the phylogenetic structuring of southern populations of L. pemnii and L. tasmaniensis. Contrarily, although the McPherson Range has previously been noted to coincide geographically with a major mtDNA phylogenetic break in Litoria fallax this pattern is not apparent in L. pemnii or L. tasmaniensis. It appears that major phylogeographic splits within L. pemnii and L. tasmaniensis may predate the Quaternary. We conclude that phylogeographies of open forest frogs are complex and more difficult to predict than for rainforest taxa, mainly due to an absence of palaeomodels for historical distributions of non-rainforest habitats.