Introduction Recent improvements in the fossil record of bats in the Americas provide new raw material to bolster interpretations of bat evolution and paleobiogeography. Herein we review the fossil record of bats from North America and South America and provide a narrative explanation of the historical development of the Neotropical bat fauna from a paleontological and biogeographical perspective. The phylogeography of bats in the Western Hemisphere as it pertains to the development of the modern Neotropical biogeographic region may also be important in developing strategies for the conservation of the tropical American chiropteran fauna. Throughout most of the Cenozoic Era there were three widely separated tropical regions in the New World: the island continent of South America, excluding temperate latitudes in Argentina and Chile; the West Indies and Middle America, including Central America and central and southern Mexico, and at times extending northward to encompass the Gulf Coastal Plain from Texas east to Florida. North America and South America possessed very different mammalian faunas prior to the Great American Biotic Interchange, which began with a limited exchange of large non-volant mammals in the Late Miocene (~9 Ma), presumably by overwater dispersal, and then became a fully fledged interchange following the formation of the Panamanian Isthmus joining these two continents in the Early Pliocene (~5 Ma). From the Pliocene onward, there was extensive intermingling between North and South American biotas, and the tropical portions of these two continents now share a common Neotropical mammalian fauna that extends northward to about the Tropic of Cancer (~23°N latitude) in central Mexico and southward to about the Tropic of Capricorn (~23°S latitude) in southern Brazil, Paraguay and northern Argentina. The tropical American bat fauna has become so extensively intermixed since the Interchange, particularly in Middle America, that it is very difficult to determine the origin of most groups that lack a fossil record. By studying pre-Pliocene bats we can establish the pre-Interchange chiropteran faunas of North and South America, thereby determining which groups of bats were involved in the Interchange and where they may have originated. The Tertiary record of New World bats is primarily limited to temperate North America and tropical South America, with two localities from Patagonia in temperate South America. The Middle American Tertiary chiropteran record is limited to the extinct vespertilionid genus †Plionycteris from two Early Pliocene sites in Mexico (Lindsay and Jacobs, 1985; Carranza-Castañeda and Walton, 1992).

The fossil history of molossids in the North American Tertiary is among the poorest for any family of bats. The oldest definite record is of Wallia scalopidens of middle Eocene (Uintan) age from Saskatchewan, Canada. One of the youngest records is of Eumops cf. E. perotis from the late Pliocene (late early Blancan) of Arizona, USA. New occurrences detailed herein from the middle and late Tertiary of Florida, USA, begin to fill in the 40-million-year gap between the previous records. They are: (1) an abraded upper molar of an indeterminate genus from the Brooksville 2 locality, Hernando County, of late Oligocene (Arikareean) age; (2) a large and a small upper molar pertaining to two congeneric species similar to Tadarida and Mormopterus, from the Thomas Farm local fauna, Gilchrist County, of early Miocene (early Hemingfordian) age; and (3) a distal humerus from a Tadarida of an unknown species that is larger than extant Tadarida brasiliensis and similarly sized but differently proportioned than in the extinct Pleistocene species Tadarida constantinei. The last specimen is from the Macasphalt Shell Pit, Sarasota County, and is of late Pliocene (late Blancan) age.

The Biodiversity Heritage Library works collaboratively to make biodiversity literature openly available to the world as part of a global biodiversity community.

The pallid bat (Antrozous pallidus) is a species of western North America, inhabiting ecoregions ranging from desert to oak and pine forest. They are primarily insectivorous predators on large arthropods that occasionally take small vertebrate prey, and are at least seasonally omnivorous in certain parts of their geographic range where they take nectar from cactus flowers and eat cactus fruit pulp and seeds. Until recently, mesquite bugs were primarily tropical-subtropical inhabitants of Mexico and Central America but have since occupied the southwestern United States where mesquite trees occur. Using a noninvasive method, we investigated the bats’ diet at the Cienega Creek Natural Preserve, Arizona, by collecting food parts discarded beneath three night roosts in soil-piping cavities in a mesquite bosque. We also made phenological and behavioral observations of mesquite bugs,Thasus neocalifornicus, and their interactions with the mesquite trees. We determined that the bats discarded inedible parts of 36 species in 8 orders of mainly large-bodied and nocturnal insects below the night-roosts. In addition, one partial bat wing represents probable predation upon a phyllostomid bat,Choeronycteris mexicana. About 17 of the insect taxa are newly reported as prey for pallid bats, as is the batC. mexicana. The majority of culled insect parts (88%) were from adult mesquite bugs. Mesquite bug nymphs did not appear in the culled insect parts. After breeding in late summer, when nighttime low temperatures dropped below 21 °C, the adult bugs became immobile on the periphery of trees where they probably make easy prey for opportunistic foliage-gleaning pallid bats. Proximity of night-roosts to mesquite bug habitat probably also enhances the bats’ exploitation of these insects in this location.

A small, late Quaternary deposit of degraded bat guano in Arkenstone Cave, Colossal Cave Mountain Park, Pima County, Arizona, yielded thousands of fossil bat bones. The bones primarily represent an attritional accumulation of dead bats beneath a maternity colony of Myotis thysanodes, a species that no longer roosts at the elevation of the cave. Rarer bones in the deposit represent a smaller species of Myotis and the extinct vampire Desmodus stocki. Due to leaching in the alkaline cave environment, the bones could not be dated by radiocarbon, but the fossils probably date to the late Pleistocene. Myotis thysanodes, a species that frequently forages in the desert but typically day-roosts at higher elevations in pygmy conifer woodland or higher vegetative zones, is consistent with the occurrence of conifer woodland at similar elevations in nearby mountain ranges in the late Pleistocene. This is the first record of D. stocki in what is now the Sonoran Desert and the second locality of occurrence for ... Resumen Miles de huesos de murciélagos fueron recuperados de un depósito pequeño de guano degradado de murciélago del Cuaternario tardío, en la Cueva Arkenstone, Colossal Cave Mountain Park, condado de Pima, Arizona. La mayoría de los huesos representan una acumulación de murciélagos muertos debajo de una colonia maternal de Myotis thysanodes, especie que ya no se refugia en la altura de la cueva. Otras especies del depósito incluyen huesos raros de una especie más pequeña de Myotis y del murciélago vampiro extinto, Desmodus stocki. La datación de los huesos con carbono radiactivo no fue posible debido a la lixiviación en el ambiente alcalino de la cueva, pero probablemente los fósiles datan del Pleistoceno tardío. La presencia de M. thysanodes, especie que frecuentemente forrajea en el desierto pero que típicamente reside a mayor altitud en bosques de coníferas pigmeas o en zonas vegetales más altas, está consistente con la presencia de bosques coníferos en alturas similares en las sierras cercanas duran...