Taxonomy and evolution
|Internal relationships of African Pteropodidae based on combined evidence of mitochondrial and nuclear DNA. One species each of Pteropodinae, Nyctimeninae, and Cynopterinae, which are not found in Africa, were included as outgroups.
The family Pteropodidae was first described in 1821 by British zoologist John Edward Gray. He named the family "Pteropidae" (after the genus Pteropus) and placed it within the now-defunct order Fructivorae. Fructivorae contained one other family, the now-defunct Cephalotidae, containing one genus, Cephalotes (now recognized as a synonym of Dobsonia). Gray's spelling was possibly based on a misunderstanding of the suffix of "Pteropus". "Pteropus" comes from Ancient Greek "pterón" meaning "wing" and "poús" meaning "foot". The Greek word pous of Pteropus is from the stem word pod-; therefore, Latinizing Pteropus correctly results in the prefix "Pteropod-".:230 French biologist Charles Lucien Bonaparte was the first to use the corrected spelling Pteropodidae in 1838.:230
In 1875, Irish zoologist George Edward Dobson was the first to split the order Chiroptera (bats) into two suborders: Megachiroptera (sometimes listed as Macrochiroptera) and Microchiroptera, which are commonly abbreviated to megabats and microbats. Dobson selected these names to allude to the body size differences of the two groups, with many fruit-eating bats being larger than insect-eating bats. Pteropodidae was the only family he included within Megachiroptera.
A 2001 study found that the dichotomy of megabats and microbats did not accurately reflect their evolutionary relationships. Instead of Megachiroptera and Microchiroptera, the study's authors proposed the new suborders Yinpterochiroptera and Yangochiroptera. This classification scheme has been verified several times subsequently and remains widely supported as of 2019. Since 2005, this suborder has alternatively been called "Pteropodiformes".:520–521 Yinpterochiroptera contained species formerly included in Megachiroptera (all of Pteropodidae), as well as several families formerly included in Microchiroptera: Megadermatidae, Rhinolophidae, Nycteridae, Craseonycteridae, and Rhinopomatidae. Two superfamilies comprise Yinpterochiroptera: Rhinolophoidea—containing the above families formerly in Microchiroptera—and Pteropodoidea, which only contains Pteropodidae.
In 1917, Danish mammalogist Knud Andersen divided Pteropodidae into three subfamilies: Macroglossinae, Pteropinae (corrected to Pteropodinae), and Harpyionycterinae.:496 A 1995 study found that Macroglossinae as previously defined, containing the genera Eonycteris, Notopteris, Macroglossus, Syconycteris, Melonycteris, and Megaloglossus, was paraphyletic, meaning that the subfamily did not group all the descendants of a common ancestor.:214 Subsequent publications consider Macroglossini as a tribe within Pteropodinae that contains only Macroglossus and Syconycteris. Eonycteris and Melonycteris are within other tribes in Pteropodinae, Megaloglossus was placed in the tribe Myonycterini of the subfamily Rousettinae, and Notopteris is of uncertain placement.
Other subfamilies and tribes within Pteropodidae have also undergone changes since Andersen's 1917 publication. In 1997, the pteropodids were classified into six subfamilies and nine tribes based on their morphology, or physical characteristics. A 2011 genetic study concluded that some of these subfamilies were paraphyletic and therefore they did not accurately depict the relationships among megabat species. Three of the subfamilies proposed in 1997 based on morphology received support: Cynopterinae, Harpyionycterinae, and Nyctimeninae. The other three clades recovered in this study consisted of Macroglossini, Epomophorinae + Rousettini, and Pteropodini + Melonycteris. A 2016 genetic study focused only on African pteropodids (Harpyionycterinae, Rousettinae, and Epomophorinae) also challenged the 1997 classification. All species formerly included in Epomophorinae were moved to Rousettinae, which was subdivided into additional tribes. The genus Eidolon, formerly in the tribe Rousettini of Rousettinae, was moved to its own subfamily, Eidolinae.
In 1984, an additional pteropodid subfamily, Propottininae, was proposed, representing one extinct species described from a fossil discovered in Africa, Propotto leakeyi. In 2018 the fossils were reexamined and determined to represent a lemur. As of 2018, there were 197 described species of megabat, around a third of which are flying foxes of the genus Pteropus.
Fossil record and divergence times
The fossil record for pteropodid bats is the most incomplete of any bat family. Several factors could explain why so few pteropodid fossils have been discovered: tropical regions where their fossils might be found are undersampled relative to Europe and North America; conditions for fossilization are poor in the tropics, which could lead to fewer fossils overall; and fossils may have been created, but they may have been destroyed by subsequent geological activity. It is estimated that more than 98% of pteropodid fossil history is missing. Even without fossils, the age and divergence times of the family can still be estimated by using computational phylogenetics. Pteropodidae split from the superfamily
Rhinolophoidea (which contains all the other families of the suborder Yinpterochiroptera) approximately 58 Mya (million years ago). The ancestor of the crown group of Pteropodidae, or all living species, lived approximately 31 Mya.
Melanesia, where many megabat subfamilies are likely to have originated
The family Pteropodidae likely originated in Australasia based on biogeographic reconstructions. Other biogeographic analyses have suggested that the Melanesian Islands, including New Guinea, are a plausible candidate for the origin of most megabat subfamilies, with the exception of Cynopterinae; the cynopterines likely originated on the Sunda Shelf based on results of a Weighted Ancestral Area Analysis of six nuclear and mitochondrial genes. From these regions, pteropodids colonized other areas, including continental Asia and Africa. Megabats reached Africa in at least four distinct events. The four proposed events are represented by (1) Scotonycteris, (2) Rousettus, (3) Scotonycterini, and (4) the "endemic Africa clade", which includes Stenonycterini, Plerotini, Myonycterini, and Epomophorini, according to a 2016 study. It is unknown when megabats reached Africa, but several tribes (Scotonycterini, Stenonycterini, Plerotini, Myonycterini, and Epomophorini) were present by the Late Miocene. How megabats reached Africa is also unknown. It has been proposed that they could have arrived via the Middle East before it became more arid at the end of the Miocene. Conversely, they could have reached the continent via the Gomphotherium land bridge, which connected Africa and the Arabian Peninsula to Eurasia. The genus Pteropus (flying foxes), which is not found on mainland Africa, is proposed to have dispersed from Melanesia via island hopping across the Indian Ocean; this is less likely for other megabat genera, which have smaller body sizes and thus have more limited flight capabilities.
Megabats are the only family of bats incapable of laryngeal echolocation. It is unclear whether the common ancestor of all bats was capable of echolocation, and thus echolocation was lost in the megabat lineage, or multiple bat lineages independently evolved the ability to echolocate (the superfamily
Rhinolophoidea and the suborder Yangochiroptera). This unknown element of bat evolution has been called a "grand challenge in biology". A 2017 study of bat ontogeny (embryonic development) found evidence that megabat embryos at first have large, developed cochlea similar to echolocating microbats, though at birth they have small cochlea similar to non-echolocating mammals. This evidence supports that laryngeal echolocation evolved once among bats, and was lost in pteropodids, rather than evolving twice independently. Megabats in the genus Rousettus are capable of primitive echolocation through clicking their tongues. Some species—the cave nectar bat (Eonycteris spelaea), lesser short-nosed fruit bat (Cynopterus brachyotis), and the long-tongued fruit bat (Macroglossus sobrinus)— have been shown to create clicks similar to those of echolocating bats using their wings.
Both echolocation and flight are energetically expensive processes. Echolocating bats couple sound production with the mechanisms engaged for flight, allowing them to reduce the additional energy burden of echolocation. Instead of pressurizing a bolus of air for the production of sound, laryngeally echolocating bats likely use the force of the downbeat of their wings to pressurize the air, cutting energetic costs by synchronizing wingbeats and echolocation. The loss of echolocation (or conversely, the lack of its evolution) may be due to the uncoupling of flight and echolocation in megabats. The larger average body size of megabats compared to echolocating bats suggests a larger body size disrupts the flight-echolocation coupling and made echolocation too energetically expensive to be conserved in megabats.
List of genera
The family Pteropodidae is divided into six subfamilies represented by 46 genera: