Overexploitation has driven catastrophic declines in most turtle species in Asia. Many turtles are seized annually by law enforcement officers; unfortunately, seized turtles often lack associated information on their provenance. There is also taxonomic confusion regarding the clear delineation of turtle species and subspecies in Asia. This lack of provenance data and taxonomic uncertainty is problematic as it may preclude release, make appropriate release sites hard to select, or result in genetic pollution when animals are released. In this study, we amassed and analyzed the largest number of genetic samples of the critically endangered big-headed turtle (Platysternon megacephalum) to date. Our phylogenetic and network results based on nearly 500 sequences of a mitochondrial ND4 fragment corroborate the earlier hypothesis that only two of the three currently recognized subspecies have strong molecular support and that there is greater genetic structuring within one subspecies than has been previously reported. The vast majority of P. megacephalum confiscated from the trade in Vietnam belong to Subclade 3 of P. m. peguense, and this subclade is the most broadly distributed taxon of the big-headed turtle in Vietnam based on samples with known localities. Nonetheless, for the other two subclades, all their samples originate from the trade and could not be assigned to any geographic provenance because of the lack of wild-caught samples. In addition, our results reveal two trade centers of the species, Quang Ninh Province on the border between China and Vietnam and Kon Tum Province in the Central Highlands. However, smaller animal seizures occur across the range of the species in Vietnam. While the data presented here have facilitated the selection of appropriate release sites for confiscated P. megacephalum, the significant costs of genetic screening and the shortage of samples from wild animals with good localities hinder the rewilding efforts, especially for conservation practitioners working in countries where conservation activities are not well funded and research on poorly studied species is still limited. Nevertheless, we urge that wildlife translocations include robust pre-translocation genetic screening to ensure that all individual animals are reintroduced at appropriate sites.

The critically endangered Vietnamese Pond Turtle (Mauremys annamensis), an endemic species occurring in a small lowland area in central Vietnam, has been virtually extirpated from its natural range. As a result, reintroduction of held individuals worldwide, especially from Europe, the United States, and Vietnam, will play a vital role in species recovery programs. Nevertheless, the discordance between different molecular markers in inferring the placements of two closely related species, M. annamensis and M. mutica, and the existence of two distinct mitochondrial lineages within M. annamensis, have hindered conservation strategies to properly maintain the genetic integrity of held populations to release individuals back to their natural habitat in the future. In this study, we sequenced 732 bps of a mitochondrial gene (ND4) and 1038 bps of an intron fragment of the RNA fingerprint protein 35 (R35) gene for 18 samples collected from the local trade in five provinces in Vietnam and 20 samples from founders in the Turtle Conservation Centre, Cuc Phuong National Park, Vietnam. DNA sequences analyzed by Bayesian Inference, Maximum Likelihood, and NeighborNet methods show that the Vietnamese Pond Turtle is a well-defined species and that the population of M. mutica from Hainan Island likely evolved through introgression between a lineage distantly related to true M. mutica and M. annamensis and a lineage closely related to the two mitochondrial haplotypes of M. annamensis. Interestingly, our analyses of samples collected from the local trade and those with known localities reveal that the two extant mitochondrial lineages within M. annamensis are geographically isolated and evolutionarily significant. Based upon the new findings, we recommend genetic screening for all held populations in Vietnam and elsewhere and that animals with different genetic histories be managed separately. In addition, it is important to use other molecular markers, such as microsatellites and single nucleotide polymorphisms (SNPs), to determine potential hybrids between the two mitochondrial haplotypes and remove them from conservation breeding programs before releasing the turtles back to their natural habitat.