Repeated evolution of underwater rebreathing in diving Anolis lizards.

Air-based respiration limits the use of aquatic environments by ancestrally terrestrial animals. To overcome this challenge, diving arthropods have evolved to respire without resurfacing using air held between their cuticle and surrounding water.1-4 Inspired by natural history observations in Haiti (unpublished data) and Costa Rica,5,6 we conducted experiments documenting routine air-based underwater respiration in several distantly related semi-aquatic Anolis lizard species. Semi-aquatic anoles live along neotropical streams and frequently dive for refuge or food,7-12 remaining underwater for up to 18 min. While submerged, these lizards iteratively expire and re-inspire narial air bubbles-underwater "rebreathing." Rebreathed air is used in respiration, as the partial pressure of oxygen in the bubbles decreases with experimental submersion time in living anoles, but not in mechanical controls. Non-aquatic anoles occasionally rebreathe when submerged but exhibit more rudimentary rebreathing behaviors. Anole rebreathing is facilitated by a thin air layer (i.e., a "plastron," sensu Brocher13) supported by the animal's rugose skin upon submergence. We suggest that hydrophobic skin, which we observed in all sampled anoles,14,15 may have been exaptative, facilitating the repeated evolution of specialized rebreathing in species that regularly dive. Phylogenetic analyses strongly suggest that specialized rebreathing is adaptive for semi-aquatic habitat specialists. Air-based rebreathing may enhance dive performance by incorporating dead space air from the buccal cavity or plastron into the lungs, facilitating clearance of carbon dioxide, or allowing uptake of oxygen from surrounding water (i.e., a "physical gill" mechanism4,16).

Past, present, and future of a freshwater fish metapopulation in a threatened landscape.

It is well documented that hydropower plants can affect the dynamics of fish populations through landscape alterations and the creation of new barriers. Less emphasis has been placed on the examination of the genetic consequences for fish populations of the construction of dams. The relatively few studies that focus on genetics often do not consider colonization history and even fewer tend to use this information for conservation purposes. As a case study, we used a 3-pronged approach to study the influence of historical processes, contemporary landscape features, and potential future anthropogenic changes in landscape on the genetic diversity of a fish metapopulation. Our goal was to identify the metapopulation's main attributes, detect priority areas for conservation, and assess the consequences of the construction of hydropower plants for the persistence of the metapopulation. We used microsatellite markers and coalescent approaches to examine historical colonization processes, traditional population genetics, and simulations of future populations under alternate scenarios of population size reduction and gene flow. Historical gene flow appeared to have declined relatively recently and contemporary populations appeared highly susceptible to changes in landscape. Gene flow is critical for population persistence. We found that hydropower plants could lead to a rapid reduction in number of alleles and to population extirpation 50-80 years after their construction. More generally, our 3-pronged approach for the analyses of empirical genetic data can provide policy makers with information on the potential impacts of landscape changes and thus lead to more robust conservation efforts.