We conducted a phylogeographic analysis of the strictly aquatic and critically endangered Central American river turtle, Dermatemys mawii, as part of a conservation management program for the species. We sampled 238 individuals from 15 different localities throughout the species range. Using sequence fragments from the mtDNA Cyt b and ND4 genes, we identified 16 different haplotypes. Overall, our results reveal a signal of phylogeographic structure throughout the range, which appears to have been secondarily blurred by extensive gene flow. Notably, this also applies to genetic structuring across three major hydrological basins that pose biogeographic breaks in other aquatic taxa. Divergence times of mtDNA haplotypes in D. mawii suggest that the main lineages split in the Pliocene-Pleistocene (3.73-0.227 MA) and demographic tests indicate that the species has undergone drastic demographic size fluctuations since this time period. One ancient haplotype (1D) was found to exhibit sequence divergence of up to 2% from other haplogroups. Divergence of this magnitude is indicative of species level differentiation in other turtle genera. Haplotype 1D was found in only two localities, Sarstun and Salinas, but specimens with other haplotypes were also found in those localities. It is not known whether the individuals with the 1D haplotype interbreed with non-1D individuals. Our results suggest that human activity, such as harvesting and long distance transport of animals, may have influenced the current patterns of genetic diversity. For more than 2000 years, D. mawii has been consumed by people from Middle American cultures, and the archeological record contains strong evidence that the Mayans transported animals between villages and far away from their natural distribution range. Therefore, the large-scale pattern of haplotype sharing even across hydrological barriers, the observed low haplotype diversity in some populations and the contemporary absence of a pronounced phylogeographic pattern is likely due to a combination of population expansions, gene flow, extensive human-mediated-movements and recent bottlenecks resulting from over-harvesting.

No abstract

suMMaRy. -The Central American River Turtle, Dermatemys mawii, the single remaining extant species of the formerly wide-spread primitive family Dermatemydidae, is restricted to the Atlantic drainages of southern Mexico, Belize, and Guatemala. This large, highly aquatic, and totally herbivorous freshwater turtle is severely threatened throughout its range. There are older accounts citing specimens over 20 kg and carapace length of 60 cm, but recent field studies have found few individuals in excess of 14 kg in Mexico or above 11 kg in Guatemala. The turtle's large size and highly palatable meat yield a high profit when the animals are sold at market. This has motivated hunters to overharvest many populations into extirpation. Although Dermatemys is consumed year-round, exploitation peaks during dry weather when low water levels facilitate capture. All forms of capture tend to select for larger size classes, thus not only reducing densities, but altering population structures, resulting in lower proportions of reproductively mature animals. The maturing cohort of smaller juveniles that remains is usually soon removed as well, resulting in unproductive populations in decline. Since Dermatemys is an herbivorous species and grows relatively fast, there is the possibility of raising this species in captivity or semicaptivity in polyculture systems with freshwater shrimp or ornamental fish. Even fast turtle growth rates would probably translate to a relatively slow cash return, suggesting that Dermatemys perhaps be a complementary species in a polyculture system. Where the capital and interest are sufficient, detailed and well-documented experiments testing the overall feasibility of these schemes are advised. However, captive propagation efforts should not distract critical funds and attention from the maintenance of wild stocks in their native habitats. Law enforcement is not the aswer to maintaining wild stocks, and habitat destruction by itself is not the problem-unless local people change their ways and stop eating or selling every Dermatemys they can, there is little chance of rehabilitating wild populations.

Natural history collections are essential to a wide variety of studies in biology because they maintain large collections of specimens and associated data, including genetic material (e.g., tissues) for DNA sequence data, yet they are currently under-funded and collection staff have high workloads. With the advent of aggregate databases and advances in sequencing technologies, there is an increased demand on collection staff for access to tissue samples and associated data. Scientists are rapidly developing large DNA barcode libraries, DNA sequences of specific genes for species across the tree of life, in order to document and conserve biodiversity. In doing so, mistakes are made. For instance, inconsistent taxonomic information is commonly taken from different lending institutions and deposited in data repositories, such as the Barcode of Life Database (BOLD) and GenBank, despite explicit disclaimers regarding the need for taxonomic verification by the lending institutions. Such errors can have profound effects on subsequent research based on these mis-labelled sequences in data repositories. Here, we present the production of a large DNA barcode library of reptiles from the National Museum of Natural History tissue holdings. The library contains 2,758 sequences (2,205 COI and 553 16S) from 2260 specimens (four crocodilians, 37 turtles, and 2,219 lizards, including snakes), representing 583 named species, from 52 countries. In generating this library, we noticed several common mistakes made by scientists depositing DNA barcode data in public repositories (e.g., BOLD and GenBank). Our goal is to raise awareness of these concerns and offer advice to avoid such mistakes in the future to maintain accurate DNA barcode libraries to properly document Earth’s biodiversity.