Bats, the order Chiroptera, are among the most abundant and diverse groups of living mammals. They occupy a wide range of ecological niches in almost all environments on Earth and are highly specialised for a vast variety of different diets and habitats. There is now in excess of 1200 species of bat, and this number is growing constantly with the discovery of new species as well as the division of those already named thanks to new genetic data. Bats are the only mammals to have evolved true flight and one of only two mammalian orders that echolocation has evolved in. The evolutionary origin of bats is still somewhat unknown due to the limited fossil record of their ancestry, and their phylogeny has been, and still is, widely debated by biologists. Recent molecular studies however, have shed more light on the early radiation of Chiroptera, and this essay will discuss the current consensus of the scientific community on the matter. We will also explore the fossil record in order to ascertain the most likely direct ancestors of the order, as well as looking at the adaptive radiation and specialisation of extant groups.
Traditionally, bats have been classified into two sub-orders, these being Megachiroptera and Microchiroptera. Although it has now been recognised that these are not actually true phylogenetic groups, they still represent the two major types of bats based on morphological features alone, and so a definition of these terms is still somewhat useful in that respect. Megachiroptera, often referred to as fruit bats, are unable to echolocate (with the exception of one species which will be mentioned later in the text) and are either nectarivorous or frugivorous, as the name suggests. Microchiroptera are often called echolocating bats, as unlike fruit bats they can generate ultrasonar calls for echolocation, or insectivorous bats, although many feed on other food sources rather than insects. Bats have long been considered to be monophyletic, however Pettigrew (1986) challenged this when he proposed that Megachiroptera did not form a sister group with Microchiroptera, because of differences in their neuroanatomy. He instead advocated that the fruit bats were more closely related to primates than they were to other bats. Although published in a leading journal, this was much disputed due to the fact that if true, Pettigrew’s ‘Flying Primate Hypothesis’ would mean that flight had evolved independently in these two groups but in no other mammals. This hypothesis continued to be the subject of much debate for the rest of the 20th century, but it is now generally agreed that the evidence for bat monophyly outweighs that for Pettigrew’s hypothesis. However, molecular studies have shown that although Chiroptera form a monophyletic group, microbats are paraphyletic. Genetic evidence suggests that the superfamily Rhinolophidea (horseshoe bats and similar species), which were previously included in the Microchiroptera, are in fact polyphyletic (Teeling et al. 2002). Studies show that some rhinolophoid families in fact more closely related to the fruit bats (Pteropodidae) than they are to other bats, with which they share the ability to echolocate with (Springer et al. 2001; Teeling et al. 2005). This has led to two new groups being proposed: the Yinpterochiroptera, consisting of fruit bats and three families of echolocating bats (Rhinolophidae, Rhinopomatidae and Megadermatidae), and the Yangochiroptera, consisting of all other bat families (Teeling et al. 2002). These new groupings can be more easily visualised in Figure 1.
Figure 1. Adapted from Teeling et al. 2005.
Phylogenetic tree showing the molecular time scale of the evolutionary radiation of order Chiroptera. The x-axis shows time (m.y.a.). Circles indicate the age of the oldest specimen in the fossil record that represents that particular lineage. ‘K-T boundary’ refers to the approximate time of the Cretaceous-Paleogene, associated with the K-T mass extinction event.
The generally most accepted time scale for the evolution of bat families was proposed by Teeling et al. 2005, as shown in Figure 1. The study used both molecular and morphological data, and showed the four major insectivorous bat lineages (Rhinolophidea, Emballonuroidea, Noctilionoidea and Vespertilionoidea) appearing between 50 and 52 million years ago. This diversification coincided with a global rise in temperature and an increase in insect diversity, which may have resulted in the diversification of these lineages due to the new food sources available to them (Teeling et al. 2005).
It is likely that the direct ancestors of bats were small arboreal insectivores, nocturnal and lacking the ability of flight (Gunnell & Simmons 2005). Molecular phylogenies support the placement of bats within Laurasiatheria (Pumo et al. 1998), the Northern superorder of mammals, suggesting that they aren’t as closely related to primates and treeshrews as morphological evidence once led us to believe. However, genetic evidence has not yet been able to determine the true sister group of bats, with Eulipotyphlans (shrews, moles and hedgehogs) or Cetferungulates (cetaceans, ungulates and carnivora) being the two major propositions for this (Van Den Bussche & Hoofer 2004; Nishihara et al. 2006). The fossil record of Chiroptera is limited compared to other mammalian orders as bat skeletons are fragile and do not preserve well. In addition to this, most specimens that we do have are not old enough to tell us anything about the origins of the order, and therefore the early evolution of bats is not yet well understood. The oldest bat fossils that have been found date back to the early Eocene, and by the mid-Eocene bats we know that bats had already diversified greatly, with many fossil specimens representing modern bat families (Gunnell & Simmons 2005). The two oldest fossil bats, Onychonycteris and Icaronycteris, coexisted approximately 50 million years ago, however Onychonycteris was by far the most primitive of the two. Figure 2 shows both of these fossils alongside a skeleton of the extant insectivorous bat, Myotis lucifugus for comparison.
Figure 2. A comparison of skeletons (in dorsal view) of the two earliest known bats with an extant insectivorous species.
a) Onychonycteris finneyi. Source: Simmons et al. 2008.
b) Icaronycterus index. Source: Simmons & Geisler 1998.
c) Myotis lucifugus (Little brown bat). A common extant insectivorous bat of North America. Photograph by M. A. Wilson (College of Wooster, OH, U.S.A), used with permission.
Whereas Icaronycterus and all other later fossil bats closely resemble modern insectivorous bats, Onychonycteris exhibits more primal features. Onychonycteris has limb proportions that are intermediate between those of extant bats and those of all other, non-flying, mammals, suggesting it was more agile and not capable of as powerful flight as its modern counterparts (Simmons et al. 2008). Onychonycteris also possessed claws on all five of its forelimb digits, rather than the maximum of two claws we see in all other extinct and modern bats, so it was likely an intermediary stage between these bats and their terrestrial predecessors (Simmons 2008). Debate is still rife as to whether flight or echolocation evolved first in bats, and Onychonycteris seemed to provide the answer to this when first found, as its ear morphology including its relatively small cochlea provided evidence for the assumption that it could not echolocate (Simmons et al. 2008; Teeling 2009). However, Veselka et al. 2010 disputed this, suggesting a way that Oynchonycteris could have echolocated and also noting that the specimen in question had been flattened during fossilisation, meaning that the exact structure of the skull could not be determined for definite.
Although it is not known whether flight or echolocation evolved first, or even if the evolution of both features coincided, biologists are fairly unanimous when it comes to considering the reasons as to why bats evolved them. Ancestors of bats are thought to have had membranes between their limbs, such as the ones in modern flying squirrels, which enabled them to glide before they developed fully functioning wings. As these early mammals were arboreal, gliding would have been advantageous because it would have conserved some of their energy when moving between trees, as well as enabling them to avoid terrestrial predators. The evolution of the bat wing seems to have happened very rapidly, and it has been associated with a single protein’s (Bmp2) expression (Sears et al. 2006). The increased agility (as seen in Onychonycteris) and the new gliding abilities of these early nocturnal ancestors of bats would have meant that their orientation skills needed to be improved in order for them to successfully hunt in the night sky. It is highly probable that echolocation in bats became increasingly more sophisticated as they became better at flying (Altringham 2011). As previously mentioned, it is now widely accepted that bats are indeed a monophyletic group, but this leads us to two different hypotheses about the evolution of echolocation within this group. Either echolocation evolved in both the Yangochiroptera and the Rhinolophidae, or it had one origin in bats and was then lost in the Pteropodidae (see Figure 1). Regardless of which is true, echolocation then evolved in the Egyptian fruit bat (Rousettus aegyptiacus), however as a less sophisticated process using tongue-clicks to produce ultrasound calls, rather than the larynx as all other echolocating bats do (Jones & Teeling 2006).
Extant bat species have diversified greatly, and whereas the single family of fruit bats, Pteropodidae, are restricted to tropical Asia and Africa, echolocating bats occur in all habitats except for the polar regions. Evolving the ability to echolocate and therefore detect and track prey in complete darkness has provided bats with the opportunity to fill the largely unoccupied niche of the night sky. Pteropodidae are exclusively herbivorous, but echolocating bats are far more diverse in their feeding habits, deriving their nourishment from many different sources such as nectar, pollen, fruit, small vertebrates, and even blood, as well as the obvious insects. It is likely that these variations on food sources originated from early bats displaying different foraging strategies, just as their modern relatives do now. As well as catching prey in the air, they probably also gleaned insects from surfaces such as plants and fruit. Exploitation of the resources near to these surfaces (e.g. nectar, pollen, and fruit) may have eventually led to specialisation and the switching of food sources in some species (Altringham 2011). Many echolocating bats are highly specialised for a particular food source, however an example of a bat species with a particularly derived feature is Desmodus rotundus, the common vampire bat. Vampire bats feed on mammalian blood, and this species has been found to have heat receptors in its nose which help it locate the areas of skin where blood flows closest too, enabling more efficient feeding (Schafer et al. 1988).
The most diverse family within the Chiroptera are the leaf nosed bats (Phyllostomidae) This is likely to be at least partly due to the fact that Phyllostomids evolved in the neotropics, which are by far the largest forests in the world, meaning that more space was available for many more species to be supported within. Both the neotropics and the Old World tropics are located close to the equator, and this may have been a contributing factor of the radiation of bat species which originated here, as species diversity is known to increase with distance to the equator (Willis & Whittaker 2002).
As we have seen in this essay, bats form the most diverse, specialised, and arguably interesting mammalian order extant today. The vast range of ecological niches they occupy no doubt contribute to their abundance all over the world and their success that has spanned at least the last 50 million years. The early evolution of Chiroptera is still poorly understood due to the limited fossil record; however, the increasing ability for molecular techniques to be employed in order to study genetic data as well as morphological features, is greatly enhancing our understanding of the adaptive radiation of bats.