Bat echolocation calls: adaptation and convergent evolution

Abstract

Bat echolocation calls provide remarkable examples of 'good design' through evolution by natural selection. Theory developed from acoustics and sonar engineering permits a strong predictive basis for understanding echolocation performance. Call features, such as frequency, bandwidth, duration and pulse interval are all related to ecological niche. Recent technological breakthroughs have aided our understanding of adaptive aspects of call design in free-living bats. Stereo videogrammetry, laser scanning of habitat features and acoustic flight path tracking permit reconstruction of the flight paths of echolocating bats relative to obstacles and prey in nature. These methods show that echolocation calls are among the most intense airborne vocalizations produced by animals. Acoustic tracking has clarified how and why bats vary call structure in relation to flight speed. Bats using broadband echolocation calls adjust call design in a range-dependent manner so that nearby obstacles are localized accurately. Recent phylogenetic analyses based on gene sequences show that particular types of echolocation signals have evolved independently in several lineages of bats. Call design is often influenced more by perceptual challenges imposed by the environment than by phylogeny, and provides excellent examples of convergent evolution. Now that whole genome sequences of bats are imminent, understanding the functional genomics of echolocation will become a major challenge.

Figure 1

(a,c) Two calls of whiskered bats and (b,d) their respective horizontal flight-induced ranging error distributions. Zero ranging error is marked by a solid line labelled ‘focus’. a: Call no.1 is short and very broadband. b: This call design has a distance of focus (DOF) of 5 cm. c: Call no.2 is long and relatively narrowband. d: It has a DOF of 111 cm. Intermediate call designs have intermediate DOFs. e: Cross-section perpendicular through a flight corridor that follows a hedge. Symbols: Twenty-two individual bats' positions with respect to the hedge. Stars indicate bats using calls in a and c, respectively. Circles indicate DOF for the calls emitted closest to the cross-sectional plain (modified from Holderied et al. 2006).

Figure 2

(a) The new molecular phylogeny of bat families. Associations that were not strongly supported by at least one independent molecular study are indicated by hatched lines. The position of the newly proposed family Miniopteridae is indicated by an asterisk. The tree unites the pteropodids, which do not use laryngeal echolocation, with the echolocating superfamily Rhinolophoidea (e.g. horseshoe bats) in the clade Yinpterochiroptera. All other bats that use laryngeal echolocation are united in the clade Yangochiroptera. Modified from Jones & Teeling (2006). (b,c) Some examples of convergent evolution in signal design. In (b), schematics of signals from Pteronotus parnellii (Yangochiroptera: Mormoopidae; left call) and Rhinolophus pearsonii (Yinpterochiroptera: Rhinolophidae; right call) are shown alongside one another. Both species emit long, constant frequency calls of similar frequency, with the constant frequency portion initiated and terminated by brief broadband sweeps. In both the cases, most energy is in the second harmonic of the call. In (c), a call of Nycteris thebaica (Yangochiroptera: Nycteridae; left call) is shown next to a call from Megaderma lyra (Yinpterochiroptera: Megadermatidae; right call). Both bats emit brief, broadband multiharmonic signals and also listen for prey-generated sounds to detect and localize prey. Call reconstructions are from illustrations in Jones & Teeling (2006), from Taylor (2000) and from unpublished recordings by G. Jones (R. pearsonii).

Figure 3

Cross-section through laser scan data over a forest track together with positions at which individual bats were flying. The section is perpendicular to the direction of the forest track and the bats' flight direction. Thirty-one flight paths are colour-coded for different species. All fly centrally but most species prefer the more open lower part while Daubenton's bats (Myotis daubentonii) also fly higher in the more confined situation (from Aschoff et al. in press).