Eavesdropping by bats on the feeding buzzes of conspecifics

Abstract

Echolocation calls of most bats are emitted at high intensities and subject to eavesdropping by nearby conspecifics. Bats may be especially attentive to "feeding buzz" calls, which are emitted immediately before attack oil airborne insects and indicate the potential presence of prey in the nearby area. Although previous work has shown that some species are attracted to feeding buzzes, these studies did not provide a well-controlled test of eavesdropping, as comparisons were made between responses to natural and altered signals (e.g., forward versus backward broadcasts of calls). In this study, I assessed the importance of feeding buzzes by conducting playbacks of controlled echolocation stimuli. I presented free-flying Brazilian free-tailed bats, Tadarida brasiliensis (I. Geoffroy, 1824), with echolocation call sequences in which feeding buzz calls were either present or absent, as well as a silence control. I determined level, of bat activity by Counting the number of echolocation calls and bat passes recorded in the presence of each stimulus, and found significantly greater bat activity in response to broadcasts that contained feeding buzzes than to broadcasts without feeding buzzes or to the silence control. These results indicate that bats are especially attentive to conspecific feeding buzz calls and that eavesdropping may allow a bat to more readily locate rich patches of insect prey.

Full text

Eavesdropping by bats on the feeding buzzes of
conspecifics
E.H. Gillam
Abstract: Echolocation calls of most bats are emitted at high intensities and subject to eavesdropping by nearby conspe-
cifics. Bats may be especially attentive to ‘‘feeding buzz’’ calls, which are emitted immediately before attack on airborne
insects and indicate the potential presence of prey in the nearby area. Although previous work has shown that some spe-
cies are attracted to feeding buzzes, these studies did not provide a well-controlled test of eavesdropping, as comparisons
were made between responses to natural and altered signals (e.g., forward versus backward broadcasts of calls). In this
study, I assessed the importance of feeding buzzes by conducting playbacks of controlled echolocation stimuli. I presented
free-flying Brazilian free-tailed bats, Tadarida brasiliensis (I. Geoffroy, 1824), with echolocation call sequences in which
feeding buzz calls were either present or absent, as well as a silence control. I determined levels of bat activity by count-
ing the number of echolocation calls and bat passes recorded in the presence of each stimulus, and found significantly
greater bat activity in response to broadcasts that contained feeding buzzes than to broadcasts without feeding buzzes or to
the silence control. These results indicate that bats are especially attentive to conspecific feeding buzz calls and that eaves-
dropping may allow a bat to more readily locate rich patches of insect prey.
Re
´sume
´:Les appels d’e
´cholocation de la majorite
´des chauves-souris sont e
´mis a
`de hautes intensite
´s et sont sujets a
`
l’e
´coute des autres chauves-souris de me
ˆme espe
`ce pre
´sentes a
`proximite
´. Les chauves-souris sont sans doute particulie
`re-
ment attentives aux « bourdonnements alimentaires » qui sont e
´mis imme
´diatement avant l’attaque d’un insecte au vol et
qui indiquent la pre
´sence de proies dans le milieu environnant. Alors que des e
´tudes ante
´rieures ont montre
´que certaines
espe
`ces sont attire
´es par les bourdonnements alimentaires, ces travaux ne fournissent pas de test bien contro
ˆle
´de l’e
´coute,
puisque les comparaisons ont e
´te
´faites entre des re
´actions a
`des signaux naturels et alte
´re
´s (par ex., la rediffusion vers
l’avant et l’arrie
`re d’enregistrements d’appels). La pre
´sente e
´tude e
´value l’importance des bourdonnements alimentaires a
`
l’aide de rediffusions de stimulus contro
ˆle
´s d’e
´cholocation. Des tadarides du Bre
´sil, Tadarida brasiliensis (I. Geoffroy,
1824), ont e
´te
´confronte
´es a
`des se
´quences d’appels d’e
´cholocation contenant ou non des bourdonnements alimentaires,
ainsi qu’a
`un silence comme te
´moin. Les niveaux d’activite
´des chauves-souris ont e
´te
´mesure
´s par le de
´nombrement des
appels d’e
´cholocation et des passages en vol en re
´action a
`chacun des stimulus; l’activite
´des chauves-souris est significa-
tivement plus importante en re
´ponse a
`des rediffusions qui contiennent des bourdonnements alimentaires qu’a
`des rediffu-
sions sans ces bourdonnements alimentaires ou qu’au silence qui sert de te
´moin. Ces re
´sultats indiquent que les chauves-
souris sont particulie
`rement attentives aux bourdonnements alimentaires des chauves-souris de leur espe
`ce et que l’e
´coute
peut leur permettre de de
´couvrir plus facilement de riches taches d’insectes proies.
[Traduit par la Re
´daction]
Introduction
Food resources often occur in small, ephemeral patches
that are separated by larger areas of poor quality. Although
such an uneven distribution may increase the time an animal
must dedicate to foraging (Stephens and Krebs 1986), efforts
to locate areas of high resource density may be enhanced by
monitoring conspecific cues associated with feeding activity
(McGregor 2005). For example, the conspicuous sounds of
an agouti, Dasyprocta punctata Gray, 1842, chewing a nut
attract other agoutis to a feeding site (Smythe 1970). Such
eavesdropping on conspecific foraging cues can occur at the
feeding site (McQuoid and Galef 1992, 1993; Nieh et al.
2004), or at a colony or roost, which may serve as an ‘‘in-
formation center’’ (Wright et al. 2003; Chauvin and Thierry
2005; Ratcliffe and ter Hofstede 2005).
Several species of bats gain knowledge of food resources
by attending to the cues of conspecifics. Seba’s short-tailed
bats, Carollia perspicillata (L., 1758), alter their food pref-
erences based on olfactory cues obtained in the roost from
conspecifics that have recently fed (Ratcliffe and ter Hof-
stede 2005). Evening bats, Nycticeius humeralis (Rafin-
esque, 1818), follow successful foragers from roosts to rich
feeding areas, with bats alternating the roles of leader and
follower on subsequent trips (Wilkinson 1992). Group de-
partures from a roost have been observed in several other
species (Racey and Swift 1985; Fenton et al. 2004),
although it is often unclear if such behavior is due to passive
information transfer (Wilkinson 1992), active recruitment of
roostmates via communication calls (Wilkinson and Bough-
man 1998), or bottlenecks at the roost exit (Speakman et al.
1992).
For insectivorous bats, echolocation is a critical sensory
process for gathering detailed information about airborne
prey. A typical echolocation sequence contains three call
Received 25 January 2007. Accepted 4 June 2007. Published on
the NRC Research Press Web site at cjz.nrc.ca on 8 August
2007.
E.H. Gillam. Department of Ecology and Evolutionary Biology,
University of Tennessee, 569 Dabney Hall, Knoxville,
TN 37996-1610, USA (e-mail: egillam@utk.edu).
795
Can. J. Zool. 85: 795–801 (2007) doi:10.1139/Z07-060 #2007 NRC Canada

phases, with each phase shift characterized by a decrease in
call duration and an increase in repetition rate (Schnitzler
and Kalko 2001). Search-phase calls are emitted before
prey detection, followed by approach-phase calls that are
produced after detection and as a bat moves toward a prey
target. The final terminal-phase, or ‘‘feeding buzz’’, calls
are emitted in rapid succession immediately before insect at-
tack (Griffin 1958; Simmons et al. 1979; Moss and Surlykke
2001). Since changes in echolocation call structure reveal
information about the foraging activity of an individual, at-
tention to the signals of conspecifics may allow bats to more
readily locate rich patches of insect prey. Furthermore, the
echolocation calls of aerial-hawking bats are very intense
compared with other natural sounds (Waters and Jones
1995), particularly those above 15 kHz, and these signals
may be especially susceptible to eavesdropping. As a result,
monitoring of calls produced by nearby conspecifics poten-
tially can lead to opportunistic aggregations of bats at insect-
rich locations (Fenton et al. 1976; Bell 1980; Vaughan 1980).
Bats may be especially attracted to feeding buzzes, as
these call sequences are typically only produced during the
final moments of attack on airborne insects (Surlykke et al.
2003). Feeding buzzes are not a direct indicator of foraging
success, as bats often exhibit low rates of successful capture
(e.g., 31% in Lasiurus borealis (Mu
¨ller, 1776), Reddy and
Fenton 2003; 35% in Eptesicus nilssonii (Keyserling and
Blasius, 1839), Rydell 1998). Alternatively, these distinctive
call sequences indicate the presence of potential prey in an
area, which may be important information for bats foraging
in the near vicinity. Three previous studies have investigated
the response of bats to echolocation call sequences contain-
ing feeding buzzes. Barclay (1982) broadcast signals of
feeding Myotis lucifugus (LeConte, 1831) (which contained
search, approach, and feeding buzz calls) to free-flying con-
specifics and he found that bat activity was significantly
higher when the foraging signal was played forwards com-
pared with trials in which the signal was played backwards.
Leonard and Fenton (1984) found similar results when per-
forming a forward–backward playback experiment with the
foraging calls of Euderma maculatum (J.A. Allen, 1891)
and suggested that this species may use echolocation to reg-
ulate individual spacing within a feeding area. Balcombe
and Fenton (1988) performed the most direct test of eaves-
dropping on feeding buzz calls, showing that foraging activ-
ity of L. borealis greatly increased in the presence of
repeated conspecific feeding buzzes compared with presen-
tations of an unedited foraging sequence that contained all
three phases of bat echolocation.
The objective of this study was to further test the hypoth-
esis that bats eavesdrop on the echolocation calls of nearby
conspecifics and are especially attracted to feeding buzzes.
Although this question has been investigated in the past,
previous studies have not examined responses to controlled
and realistic playback signals in which terminal-phase calls
were either present or absent. The backward broadcasts con-
ducted by Barclay (1982) and Leonard and Fenton (1984)
altered important information about the echolocation signal,
including the direction of a call’s frequency sweep and the
temporal pattern of the call sequence (Barclay 1982). It is
possible that the altered signal was not recognized by bats
as a sequence of foraging calls, resulting in lower responses
compared with the forward broadcast, independent of the
presence or absence of feeding buzzes. Balcombe and Fen-
ton (1988) used a ‘‘super stimulus’’ of 51 repeated feeding
buzzes that did not contain the search- and approach-phase
signals which almost always occur between consecutive
buzzes (Schnitzler and Kalko 2001). In this study, I per-
formed a controlled experiment in which I compared bat ac-
tivity in response to two echolocation playback stimuli: (1) a
call sequence that contained only search-phase calls and
(2) the same search-phase sequence with a typical series of
approach-phase and feeding buzz calls added at regular in-
tervals. This design also allowed me to control for other
stimulus characteristics such as call frequency and duration.
I chose to investigate the effects of conspecific feeding
buzzes with Brazilian free-tailed bats, Tadarida brasiliensis
(I. Geoffroy, 1824). This species is highly gregarious, form-
ing colonies that reach into the millions in south-central
Texas (Davis et al. 1962). Despite their ability to disperse
25 km or more from the roost (Davis et al. 1962; Williams
et al. 1973), these bats experience a high interaction rate
with conspecifics while foraging (Ratcliffe et al. 2004). The
echolocation calls of T. brasiliensis are generally narrow-
band and relatively low in frequency (mean minimum fre-
quency of 22.3 kHz; Gillam and McCracken 2007); as a
result, calls will propagate substantial distances in the envi-
ronment and be audible to nearby bats. This foraging behav-
ior makes T. brasiliensis an optimal species for studying the
response of bats to the feeding buzzes of conspecifics.
Materials and methods
Field experiments with echolocation playbacks
I performed playback experiments between 2120 and
0020 on eight nights from 3 June to 12 June 2006. All ex-
periments were performed on a cotton farm in the vicinity
of Uvalde, Texas, which is close to several large Brazilian
free-tailed bat colonies, and bats were often observed forag-
ing on insects found in high densities over the crop fields
where I conducted the study. Bat activity during this time
period was relatively steady, generally with one to two bats
foraging in the area for a short period of time (~1 min), sep-
arated by varying periods of no activity (E.H. Gillam, per-
sonal observation).
Playback stimuli were constructed from previously ob-
tained recordings of bats foraging at the study site. The first
playback signal, or ‘‘feeding buzzes present’’, contained
calls from the search, approach, and terminal phases of bat
echolocation (Fig. 1a). I assembled this signal by repeating
one typical search-phase call at 200 ms intervals for 8.8 s
and appending a 1.45 s sequence of approach-phase and
feeding buzz calls. This 10.25 s composite sequence was re-
peated to create a 10 min playback. The second playback
signal, or ‘‘feeding buzzes absent’’, contained only search-
phase calls (Fig. 1b) and was constructed by repeating the
same search-phase call from the first playback at 200 ms in-
tervals to create a 10 min signal. I also used a 10 min con-
trol broadcast containing no sound, referred to as ‘‘silence’’.
Each night, I broadcast six replicates of each stimulus in a
mixed order and changed the playback order on successive
nights to control for temporal effects. This design ensured
an even distribution of the stimulus presentations through-
796 Can. J. Zool. Vol. 85, 2007
#2007 NRC Canada

out the evenings of the study period. I broadcast stimuli
through an omnidirectional ultrasonic speaker (Avisoft
60401, Avisoft Bioacoustics, Berlin, Germany; flat fre-
quency response ± 5 dB between 15 and 43 kHz) mounted
on a tripod 3 m from the ground. Broadcast amplitude was
assessed by converting the intensity of an equal-amplitude
continual tone to the peak equivalent sound pressure level
(peSPL; Stapells et al. 1982) with a B&K ¼ condensor micro-
phone No. 4939 and a B&K measuring amplifier No. 2606
(Bru
¨el and Kjær, Nærum, Denmark). Broadcast amplitude
was determined to be 88 dB peSPL, which is lower than
the typical call amplitude of many insectivorous bats
(>100 dB; Lawrence and Simmons 1982; Waters and Jones
1995), but was the highest amplitude possible without
overloading the speaker. A solid dielectric microphone
(Avisoft CM16; flat frequency response ±3 dB between 10
and 100 kHz) was positioned 2 m to the left of the speaker
at a height of 3 m and oriented directly upward. Stimuli
were generated from a Dell Inspiron Laptop through a
high-speed sound card (DAQCard-6062E; National Instru-
ments, Austin, Texas) and an amplifier (Avisoft 70101)
powered by three 12 V, 7.2 A gel cell batteries. High-
speed data acquisition was accomplished with Avisoft’s
Ultrasound Gate 416 through the same laptop that was
used for broadcasts. Both playback and recording were
conducted with Avisoft RECORDER. Recordings were
5 min long, but sampling was continuous because there
was no time gap between consecutive recording files. Re-
cordings were made with 16-bit resolution and a 166 kHz
sampling rate, and included both the playback signal and
the calls of free-flying bats in the area.
Measurements of bat activity
All acoustic measurements and analyses were conducted
with Avisoft-SasLab Pro. I digitally high-pass-filtered all re-
cordings to remove background noise, using a finite impulse
response filter with a 5 kHz cutoff. I excluded from analysis
files that contained high levels of wind noise. I assessed lev-
els of bat activity in the presence of the three playback sig-
nals by counting the following: (i) number of bat calls and
(ii) number of bat passes. Both of these measurements pro-
vided information about relative bat activity in the presence
of each playback stimulus, but they could not be used to es-
timate the absolute number of bats in the recording area.
A pulse-train analysis was used to automatically detect
and count echolocation calls. This analysis used a hystere-
sis searching method to detect calls, which involved count-
ing an amplitude peak only if it exceeded the pre-peak
amplitude by a pre-defined threshold (Specht 2004). The
value of this hysteresis threshold influences the pulse
counts produced by the program. To ensure that I chose an
appropriate value, I counted several recordings by hand and
compared my counts with those produced by the pulse-train
analysis at different hysteresis settings. A 20 dB hysteresis
threshold yielded the most accurate pulse counts, and thus
was used for all analyses.
The pulse-count analysis also was influenced by the am-
plitude threshold setting, with lower thresholds resulting in
increased detection of weak signals and a higher final pulse
count. I counted the calls in each recording file using three
amplitude thresholds: 100, 300, and 500 mV (Fig. 2). This
allowed me to assess the relative amplitude of detected calls
and to gain insight into how close bats were flying to the
Fig. 1. Spectrogram of the final 2 s portion of the playback stimulus sequences. (a) ‘‘Feeding buzzes present’’ stimulus containing search-,
approach-, and terminal-phase calls of Tadarida brasiliensis. The preceding 8.25 s of the call sequence that is not shown is composed of
search-phase calls that are identical to the first two signals in this shortened sequence. (b) ‘‘Feeding buzzes absent’’ stimulus containing
search-phase calls only. The preceding 8.25 s of the call sequence that is not shown is identical to the depicted calls.
Gillam 797
#2007 NRC Canada

recording system. Whereas the 100 mV analysis detected the
greatest number of calls (including weak calls from more
distant bats), the 300 and 500 mV analyses counted a de-
creasing number of pulses, only detecting higher amplitude
signals (Fig. 2).
Playback calls were present in recordings but were only
detectable by the 100 mV analysis. To obtain a pulse count
that excluded the playback calls, I broadcast each stimulus
before bats arrived at the study site and counted the number
of detected pulses in these ‘‘bat-free’’ recordings. I then sub-
tracted the appropriate playback pulse count (feeding buzzes
present or feeding buzzes absent) from the counts produced
by the 100 mV analyses. Another issue was that some bat
calls overlapped with playback calls and were not counted.
Although this led to lower pulse counts, I chose not to in-
clude a correction for this overlap error in the final analysis,
as corrections resulted in only small changes to the final
counts and the unadjusted value was more conservative.
Assessing bat activity by counting individual calls can
lead to issues of pseudoreplication, as calls emitted in a se-
quence by the same bat are not independent of each other
(Hurlbert 1984). To address this issue, I conducted a manual
count of bat passes in addition to the automated pulse-count
analyses. Since a bat pass consists of a series of adjacent
calls that are likely produced by one bat, issues of depend-
ence are substantially reduced. A bat pass was defined as a
sequence of 2 or more echolocation calls (Furlonger et al.
1987) exceeding an amplitude of 300 mV, with adjacent
passes separated by at least 1 s of silence (Seidman and Za-
bel 2001). A 300 mV amplitude threshold allowed me to ex-
clude weak calls of distant bats and to only focus on the calls
of animals in the immediate vicinity of the experimental
setup.
Statistical analyses
I tested if the number of detected bat passes and bat calls
differed between the three playback stimuli (feeding buzzes
present, feeding buzzes absent, silence) by conducting GLM
ANCOVAs and post hoc Tukey–Kramer multiple compari-
son tests. Because of the potential influence of differences
in weather conditions between recording nights, date was in-
cluded as a covariate in all analyses. A significance level of
0.05 was used for all tests.
Results
I analyzed thirty-eight 10 min recordings for each of the
three playback signals (n= 114 total). I excluded thirty
10 min recordings in which spurious noise caused by strong
winds was present and obscured the signals of calling bats.
The number of recorded bat calls was significantly different
between stimuli for detection thresholds of 100 mV
(F[2,114] = 3.21, P= 0.044), 300 mV (F[2,114] = 13.63, P<
0.0001), and 500 mV (F[2,114] = 14.94, P< 0.0001). Date
was not a significant covariate for any count analysis. For
the 300 and 500 mV analyses, the Tukey–Kramer tests re-
vealed that bat activity was significantly greater in response
to the feeding buzzes present broadcast compared with the
feeding buzzes absent or silence stimulus (Figs. 3B, 3C).
Despite a significant Pvalue for the ANCOVA test, no dif-
ferential response between broadcast stimuli was observed
for the 100 mV analysis, although there is an obvious trend
for the same pattern of increased activity in response to the
feeding buzzes present stimulus (Fig. 3A). The number of
recorded bat passes was also significantly different between
the three playback stimuli (F[2,114] = 4.83, P= 0.009), with
date as a significant covariate (P= 0.018). Mean number of
bat passes in a 10 min recording period was greatest for the
Fig. 2. Amplitude envelope depicting the detection of five calls of T. brasiliensis with inter-call intervals of approximately 210 ms. In this
example, the 100 mV amplitude threshold detects all five calls, the 300 mV detects three calls, and the 500 mV threshold only detects the
loudest call.
798 Can. J. Zool. Vol. 85, 2007
#2007 NRC Canada

feeding buzzes present stimulus and lowest for the silence
control, with an intermediate value for the feeding buzzes
absent signal (Fig. 3D).
Discussion
The results of this study support the hypothesis that bats
eavesdrop on the echolocation calls of conspecifics and are
attracted to terminal-phase feeding buzzes which indicate
the potential presence of insect prey. Although significant
differences existed between stimuli for all three pulse-count
analyses, it is interesting that the largest differences were
observed in the 300 and 500 mV analyses (Figs. 3B and
3C, respectively). This suggests that the playback stimulus
containing feeding buzzes not only attracted more bats, but
that these bats more closely approached our speaker system,
as revealed by the high amplitude of the detected echoloca-
tion calls. Furthermore, the playback amplitudes used here
were lower than the natural call amplitudes of many insec-
tivorous species (88 vs. >100 dB; Lawrence and Simmons
1982; Waters and Jones 1995), suggesting that results from
this study are likely conservative compared with eavesdrop-
ping among multiple free-flying bats. Overall, these results
provide evidence that bats approach conspecifics emitting
terminal-phase calls, likely in an attempt to enhance feeding
success or to gain more detailed information about the sig-
naling animal and its foraging area.
Our findings agree with those of previous studies demon-
strating that bats are attracted to terminal-phase feeding
buzzes (Barclay 1982; Leonard and Fenton 1984; Balcombe
and Fenton 1988). Although Barclay (1982) reported that
more bats responded to foraging calls played forward com-
pared with calls played backward, he also indicated feeding
buzzes may not be critical to eavesdropping, as there were
no differences in the response of M. lucifugus to conspecific
‘‘foraging’’ and ‘‘non-foraging’’ sequences. However, this
non-foraging playback was recorded during swarming, when
bats aggregate for mating (McCracken and Wilkinson 2000),
and could possibly have contained other signals (e.g., social
communication calls) that attracted bats to the playback de-
spite the absence of feeding buzzes. Here I demonstrate that
when exposed to two realistic and otherwise identical echo-
location stimuli, the signal containing approach-phase and
feeding buzz calls was more attractive, suggesting that bats
pay particular attention to the portion of a call sequence
which is associated with insect attack.
Fig. 3. Mean activity of T. brasiliensis during a 10 min broadcast of each playback stimulus. All ANCOVA tests were significant at the 0.05
level. Letters indicate results of the Tukey–Kramer multiple comparison tests; counts from stimuli labeled aare not significantly different
from each other, but are significantly different from stimuli labeled b. (A–C) Mean (±SE) number of bat calls as detected by the automated
pulse-count analyses using amplitude thresholds of 100 mV (A), 300 mV (B), and 500 mV (C). (D) Mean (±SE) number of bat passes
detected by manual counts of call sequences containing two or more echolocation pulses that exceeded an amplitude threshold of 300 mV.
Gillam 799
#2007 NRC Canada

The question remains as to whether eavesdropping on
conspecific feeding buzzes represents information parasitism
or information transfer. Information parasitism occurs if for-
aging success decreases when conspecifics are attracted to a
bat’s foraging area, whereas information transfer occurs if
sharing knowledge of foraging spots either does not affect
or increases an individual’s foraging success (Wilkinson
1992, 1995). Since eavesdropping is a passive process by
which individuals attend and respond to the foraging cues
of conspecifics, either information parasitism or information
transfer could be occurring.
The incidence of aggressive territorial behaviors among
conspecifics may indicate if the aggregation of bats at a
feeding sight owing to eavesdropping affects individual for-
aging success. Reddy and Fenton (2003) reported that
L. borealis exhibit aggressive, intraspecific chases at high
bat densities, although this behavior may not be associated
with lower prey densities (Hickey and Fenton 1990). Racey
and Swift (1985) found that aggressive interactions between
foraging Pipistrellus pipistrellus (Schreber, 1774) were high-
est when insect density at the feeding site was low. Alterna-
tively, at very high insect densities, >30 bats foraged in a
small space and showed no aggressive behavior, suggesting
that in P. pipistrellus aggression is primarily influenced by
insect density instead of bat density. Brazilian free-tailed
bats aggregate in large numbers and ultimately encounter
many conspecifics while foraging (Ratcliffe et al. 2004). De-
spite this high encounter rate, agonistic interactions are
rarely observed (G.F. McCracken, personal communication)
and large numbers of bats are commonly seen foraging over
crop fields in close proximity to one another. This lack of
aggression between foraging T. brasiliensis suggests that
eavesdropping on the echolocation calls of conspecifics
does not lead to reduced individual foraging success and is
best described as passive information transfer.
The rich food sources exploited by Brazilian free-tailed
bats also compliment the hypothesis of passive information
transfer. Moths are a major food source for T. brasiliensis,
sometimes comprising over 80% of their diet (Whitaker et
al. 1996; Lee and McCracken 2002). Noctuid moths, such
as Helicoverpa zea (Boddie, 1850), are very abundant in
south-central Texas and their distribution is highly variable
in space and time (Fitt 1989). Mass emergences of billions
of moths occur asynchronously over crop fields within brief
time windows (Raulston et al. 1990), resulting in strong spa-
tial and temporal heterogeneities in resource availability
(J.K. Westbrook and E.H. Gillam, unpublished data (2004)).
Such rich, ephemeral patches of insects have previously
been suggested to be ideal conditions for information trans-
fer, as patches contain sufficient prey to support successful
foraging by multiple bats but do not persist long enough to
warrant territoriality and defense (Wilkinson 1992). Although
high bat densities may lead to increased acoustic interfer-
ence from the calls of nearby bats, some species appear to
forage effectively in the presence of many conspecifics
(Racey and Swift 1985). Furthermore, T. brasiliensis exhibit
jamming avoidance responses (Gillam et al. 2007), which
should reduce the amount of interference caused by the calls
of other bats. Overall, eavesdropping by Brazilian free-tailed
bats should allow individuals to enhance foraging success by
decreasing the amount of time spent in a poor area and gain-
ing information about the presence of new, ephemeral patches
that are rich in insect prey (Galef and Giraldeau 2001).
Although echolocation signals in bats are primarily used
for orientation and prey detection, it has been suggested
that echolocation evolved from social communication calls
(Fenton 1984). Most bats emit a wide range of social calls
that are associated with several behaviors, including mating
(Bradbury 1977), mother–young interactions (Balcombe and
McCracken 1992), and alarm signaling (Russ et al. 1998).
The results of this study enhance the link between echoloca-
tion and social calls by further demonstrating that echoloca-
tion can have a communicative function.
Acknowledgements
I thank Gary McCracken for help with the experimental
design and critical reading of earlier versions of the manu-
script, as well as Jim Hall for assistance with measurements
of signal intensity. This research was supported by an Envi-
ronmental Protection Agency Science to Achieve Results
Graduate Research Fellowship, a University of Tennessee
SARIF grant, National Science Foundation – Experimental
and Integrative Activities award (NSF–EIA) 0326483 (T.H.
Kunz, principal investigator; M. Betke, G.F. McCracken,
P. Morton, and J.K. Westbrook, co-principal investigators),
and a research award from the Department of Ecology and
Evolutionary Biology at the University of Tennessee.
References
Balcombe, J.P., and Fenton, M.B. 1988. Eavesdropping by bats —
the influence of echolocation call design and foraging strategy.
Ethology, 79: 158–166.
Balcombe, J.P., and McCracken, G.F. 1992. Vocal recognition in
Mexican free-tailed bats — do pups recognize mothers? Anim.
Behav. 43: 79–87. doi:10.1016/S0003-3472(05)80073-9.
Barclay, R.M.R. 1982. Interindividual use of echolocation calls —
eavesdropping by bats. Behav. Ecol. Sociobiol. 10: 271–275.
doi:10.1007/BF00302816.
Bell, G.P. 1980. Habitat use and response to patches of prey by de-
sert insectivorous bats. Can. J. Zool. 58: 1876–1883.
Bradbury, J.W. 1977. Lek mating behavior in the hammer-headed
bat. Z. Tierpsychol. 45: 225–255.
Chauvin, C., and Thierry, B. 2005. Tonkean macaques orient their
food search from olfactory cues conveyed by conspecifics.
Ethology, 111: 301–310. doi:10.1111/j.1439-0310.2004.01066.x.
Davis, R.B., Herreid, C.F., and Short, H.L. 1962. Mexican free-
tailed bats in Texas. Ecol. Monogr. 32: 311–346. doi:10.2307/
1942378.
Fenton, M.B. 1984. Echolocation: implications for ecology and
evolution of bats. Q. Rev. Biol. 59: 33–53.
Fenton, M.B., Belwood, J.J., Fullard, J.H., and Kunz, T.H. 1976.
Responses of Myotis lucifugus (Chiroptera: Vespertilionidae) to
calls of conspecifics and to other sounds. Can. J. Zool. 54:
1443–1448.
Fenton, M.B., Jacobs, D.S., Richardson, E.J., Taylor, P.J., and
White, E. 2004. Individual signatures in the frequency-modulated
sweep calls of African large-eared, free-tailed bats Otomops mar-
tiensseni (Chiroptera: Molossidae). J. Zool. (Lond.), 262: 11–19.
Fitt, G.P. 1989. The ecology of Heliothis species in relation to
agroecosytems. Annu. Rev. Entomol. 34: 17–52. doi:10.1146/
annurev.en.34.010189.000313.
Furlonger, C.L., Dewar, H.J., and Fenton, M.B. 1987. Habitat use
by foraging insectivorous bats. Can. J. Zool. 65: 284–288.
800 Can. J. Zool. Vol. 85, 2007
#2007 NRC Canada

Galef, B.G., and Giraldeau, L.A. 2001. Social influences on fora-
ging in vertebrates: causal mechanisms and adaptive functions.
Anim. Behav. 61: 3–15. doi:10.1006/anbe.2000.1557.
Gillam, E.H., and McCracken, G.F. 2007. Variability in the echolo-
cation of Tadarida brasiliensis: effects of geography and local
acoustic environment. Anim. Behav. In press. PMID:17389908.
Gillam, E.H., Ulanovsky, N., and McCracken, G.F. 2007. Rapid
jamming avoidance in biosonar. Proc. R. Soc. Lond. B Biol.
Sci. 274: 651–660. doi:10.1098/rspb.2006.0047.
Griffin, D.R. 1958. Listening in the dark. Yale University Press,
New Haven, Conn.
Hickey, M.B.C., and Fenton, M.B. 1990. Foraging by red bats (La-
siurus borealis): Do intraspecific chases mean territoriality?
Can. J. Zool. 68: 2477–2482.
Hurlbert, S.H. 1984. Pseudoreplication and the design of ecological
field experiments. Ecol. Monogr. 54: 187–211. doi:10.2307/
1942661.
Lawrence, B.D., and Simmons, J.A. 1982. Measurements of atmo-
spheric attenuation at ultrasonic frequencies and the significance
for echolocation by bats. J. Acoust. Soc. Am. 71: 585–590.
doi:10.1121/1.387529. PMID:7085967.
Lee, Y.-F., and McCracken, G.F. 2002. Foraging activity and food
resource use of Brazilian free-tailed bats, Tadarida brasiliensis
(Molossidae). Ecoscience, 9: 306–313.
Leonard, M.L., and Fenton, M.B. 1984. Echolocation calls of Eu-
derma maculatum (Vespertilionidae) — use in orientation and
communication. J. Mammal. 65: 122–126. doi:10.2307/1381209.
McCracken, G.F., and Wilkinson, G.S. 2000. Bat mating systems.
In Reproductive biology of bats. Edited by E.G. Chrichton and
P.H. Krutsch. Academic Press, San Diego, Calif.
McGregor, P.K. 2005. Animal communication networks. Cam-
bridge University Press, Cambridge.
McQuoid, L.M., and Galef, B.G. 1992. Social influences on feed-
ing site selection by Burmese fowl (Gallus gallus). J. Comp.
Psychol. 106: 137–141. doi:10.1037/0735-7036.106.2.137.
McQuoid, L.M., and Galef, B.G. 1993. Social stimuli influencing
feeding behavior of Burmese fowl — a video analysis. Anim.
Behav. 46: 13–22. doi:10.1006/anbe.1993.1157.
Moss, C.F., and Surlykke, A. 2001. Auditory scene analysis by
echolocation in bats. J. Acoust. Soc. Am. 110: 2207–2226.
doi:10.1121/1.1398051. PMID:11681397.
Nieh, J.C., Barreto, L.S., Contrera, F.A.L., and Imperatriz-Fonseca,
V.L. 2004. Olfactory eavesdropping by a competitively foraging
stingless bee, Trigona spinipes. Proc. R. Soc. Lond. B Biol. Sci.
271: 1633–1640. doi:10.1098/rspb.2004.2717.
Racey, P.A., and Swift, S.M. 1985. Feeding ecology of Pipistrellus
pipistrellus (Chiroptera: Vespertilionidae) during pregnancy and
lactation. I. Foraging behavior. J. Anim. Ecol. 54: 205–215.
Ratcliffe, J.M., and ter Hofstede, H.M. 2005. Roosts as information
centres: social learning of food preferences in bats. Biol. Lett. 1:
72–74. doi:10.1098/rsbl.2004.0252. PMID:17148131.
Ratcliffe, J.M., ter Hofstede, H.M., Avila-Flores, R., Fenton, M.B.,
McCracken, G.F., Biscardi, S., Blasko, J., Gillam, E., Orprecio,
J., and Spanjer, G. 2004. Conspecifics influence call design in
the Brazilian free-tailed bat, Tadarida brasiliensis. Can. J. Zool.
82: 966–971. doi:10.1139/z04-074.
Raulston, J.R., Summy, K.R., Loera, J., Pair, S.D., and Sparks,
A.N. 1990. Population dynamics of corn earworm larvae (Lepi-
doptera: Noctuidae) on corn in the Lower Rio Grande Valley.
Environ. Entomol. 19: 274–280.
Reddy, E., and Fenton, M.B. 2003. Exploiting vulnerable prey:
moths and red bats (Lasiurus borealis; Vespertilionidae). Can.
J. Zool. 81: 1553–1560. doi:10.1139/z03-146.
Russ, J.M., Racey, P.A., and Jones, G. 1998. Intraspecific re-
sponses to distress calls of the pipistrelle bat, Pipistrellus pipis-
trellus. Anim. Behav. 55: 705–713. doi:10.1006/anbe.1997.0665.
Rydell, J. 1998. Bat defence in lekking ghost swifts (Hepialus hu-
muli), a moth without ultrasonic hearing. Proc. R. Soc. Lond. B
Biol. Sci. 265: 1373–1376.
Schnitzler, H.U., and Kalko, E.K.V. 2001. Echolocation by insect-
eating bats. Bioscience, 51: 557–569. doi:10.1641/0006-
3568(2001)051[0557:EBIEB]2.0.CO;2.
Seidman, V.M., and Zabel, C.J. 2001. Bat activity along intermit-
tent streams in northwestern California. J. Mammal. 82: 738–
747. doi:10.1644/1545-1542(2001)082<0738:BAAISI>2.0.CO;2.
Simmons, J.A., Fenton, M.B., and O’Farrell, M.J. 1979. Echoloca-
tion and pursuit of prey by bats. Science (Washington, D.C.),
203: 16–21. doi:10.1126/science.758674. PMID:758674.
Smythe, N. 1970. Ecology and behavior of the agouti (Dasyprocta
punctata) and related species on Barro Colorado Island. Ph.D.
dissertation, University of Maryland, College Park.
Speakman, J.R., Bullock, D.J., Eales, L.A., and Racey, P.A. 1992.
A problem defining temporal pattern in animal behavior —
clustering in the emergence behavior of bats from maternity
roosts. Anim. Behav. 43: 491–500. doi:10.1016/S0003-3472(05)
80107-1.
Specht, R. 2004. Avisoft-SASLab Pro: sound analysis and synthesis
laboratory. Avisoft Bioacoustics, Berlin, Germany. Available
from http://www.avisoft.com/ (accessed 3 June 2004).
Stapells, D.R., Picton, T.W., and Smith, A.D. 1982. Normal hear-
ing thresholds for clicks. J. Acoust. Soc. Am. 72: 74–79.
doi:10.1121/1.388026. PMID:7108045.
Stephens, D.W., and Krebs, J.R. 1986. Foraging theory. Princeton
University Press, Princeton, N.J.
Surlykke, A., Futtrup, V., and Tougaard, J. 2003. Prey-capture suc-
cess revealed by echolocation signals in pipistrelle bats (Pipis-
trellus pygmaeus). J. Exp. Biol. 206: 93–104. doi:10.1242/jeb.
00049. PMID:12456700.
Vaughan, T.A. 1980. Opportunistic feeding by two species of Myo-
tis. J. Mammal. 61: 118–119. doi:10.2307/1379964.
Waters, D.A., and Jones, G. 1995. Echolocation call structure and
intensity in five species of insectivorous bats. J. Exp. Biol. 198:
475–489. PMID:7699316.
Whitaker, J.O., Neefus, C., and Kunz, T.H. 1996. Dietary variation
in the Mexican free-tailed bat (Tadarida brasiliensis mexicana).
J. Mammal. 77: 716–724. doi:10.2307/1382676.
Wilkinson, G.S. 1992. Information transfer at evening bat colonies.
Anim. Behav. 44: 501–518. doi:10.1016/0003-3472(92)90059-I.
Wilkinson, G.S. 1995. Information transfer in bats. Symp. Zool.
Soc. London, 67: 345–360.
Wilkinson, G.S., and Boughman, J.W. 1998. Social calls coordinate
foraging in greater spear-nosed bats. Anim. Behav. 55: 337–350.
doi:10.1006/anbe.1997.0557.
Williams, T.C., Ireland, L.C., and Williams, J.M. 1973. High alti-
tude flights of the free-tailed bat, Tadarida brasiliensis, observed
with radar. J. Mammal. 54: 807–821. doi:10.2307/1379076.
Wright, J., Stone, R.E., and Brown, N. 2003. Communal roosts as
structured information centres in the raven, Corvus corax.
J. Anim. Ecol. 72: 1003–1014. doi:10.1046/j.1365-2656.2003.
00771.x.
Gillam 801
#2007 NRC Canada