Behavioral Ecology Vol. 11 No. 5: 528-535
© 2000 International Society for Behavioral Ecology
Insular tammar wallabies (Macropus eugenii) respond to visual but not acoustic cues from predators
a Department of Biology, Macquarie University, Sydney, NSW 2109, Australia b Department of Psychology, Macquarie University, Sydney, NSW 2109, Australia c The Cooperative Research Centre for the Conservation and Management of Marsupials, Macquarie University, Sydney, NSW 2109, Australia
Address correspondence to D. T. Blumstein. E-mail: dan{at}galliform.bhs.mq.edu.au .
Received 27 August 1999; revised 25 November 1999; accepted 30 January 2000.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
We studied the way in which a population of tammar wallabies (Macropus eugenii), which have been isolated from mammalian predators since the last ice age, responded to the sight and sound of historical and ontogenetically and evolutionarily novel predators. Tammars were shown a range of visual stimuli, including taxidermic mounts of two evolutionarily novel predators, a red fox (Vulpes vulpes) and a cat (Felis catus), and a model of an extinct predator, the thylacine (Thylacinus cynocephalus). Controls were a conspecific, the cart on which all mounts were presented, and blank trials in which spontaneous change in behavior was measured. We played back recorded sounds to characterize responses to acoustic cues from predators and to a putative conspecific antipredator signal. Treatments included the howls of dingoes (Canis lupus dingo), an evolutionarily novel predator; calls of a wedge-tailed eagle (Aquila audax), a historical and current predator; and wallaby foot thumps. Controls were the song of an Australian magpie (Gymnorhina tibicen) and a blank trial. After seeing a fox, wallabies thumped their hind feet in alarm, suppressed foraging, and increased looking. The sight of a cat similarly suppressed foraging and increased looking. The sounds of predators did not influence responsiveness, but wallabies foraged less and looked more after thump playbacks. Our results suggest that tammars respond to the sight, but not the sounds, of predators. In contrast, the response to foot thumps demonstrates that this particular sound functions as an antipredator signal. We suggest that responsiveness to visual cues has been preserved under relaxed selection because predator morphology is convergent, but vocalizations are not.
Key words: antipredator behavior, Macropus eugenii, predator recognition, relaxed selection, tammar wallaby.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The degree to which antipredator behavior persists or is lost under relaxed selection is of considerable theoretical interest (Byers, 1997
; Coss,
1991,
1999;
Magurran, 1999), and it also
has practical implications for the conservation and management of
geographically isolated populations (Berger,
1998,
1999). In some cases, species
that do not encounter predators for many generations lose their ability to
respond to them. This evolutionary isolation may be natural, as might occur on
an island isolated from the mainland, or in other predator-free habitat
patches (Curio, 1966,
1975;
Kavaliers, 1990;
Levesley and Magurran, 1988;
Seghers, 1974). Alternatively,
isolation may result from human-induced habitat fragmentation or population
management (Berger, 1998).
Antipredator behavior is not inevitably lost under such circumstances; some
species retain effective responses to predators after long periods of
isolation. The factors responsible for retention are a topic of considerable
interest (Coss, 1999;
Curio, 1966;
Pressley, 1981). For instance,
species might lose a specific response evoked by a particular predator, while
retaining more general antipredator behavior evoked by cues that are shared by
several predators. In this study, we quantified the way in which individuals
from a population isolated from some predators since the last ice age
responded to the sights and sounds of both historical and evolutionarily novel
predators. We wanted to determine whether responsiveness is retained under
relaxed selection and also whether it can be generalized to novel
predators.
In principle, prey may respond to cues associated with predators in any
sensory modality. In the present study we concentrated on the effects of
visual and acoustic stimuli generated by predators and also characterized the
response evoked by a putative acoustic antipredator signal, the foot thump
(Coulson, 1996). Visual
predator recognition sometimes depends on simple cues, such as apparent size,
shape, speed, or the presence of frontally positioned eyes (e.g.,
Coss and Goldthwaite, 1995;
Curio, 1993;
Evans et al., 1993;
Tinbergen, 1951). These
features are typically common to a class of predators (e.g., carnivores have
binocular vision and positioned eyes; attacking raptors fly quickly) and thus
permit potential prey to respond both to historical predators and to
evolutionarily novel ones (but see Hirsch
and Bolles, 1980).
Acoustic predator recognition may be different. While hunting, predators
are typically quiet and stealthy, but they must also communicate with
conspecifics (e.g., Schaller,
1972). These social signals reveal the presence of a predator and
could be exploited by omniscient prey to enhance vigilance. In fact,
antipredator behavior is reliably elicited by the sounds produced by some
species' predators (Hauser and Wrangham,
1990; Hendrie et al.,
1998; Macedonia and Yount,
1991;
Noë and
Bshary, 1997; Swaisgood et
al., 1999;
Zuberbühler
et al., 1997). Among predators, selection for species
identification leads to divergent signal structure (e.g., the territorial
vocalizations of related canids are structurally distinct;
Estes, 1991). Consequently,
acoustic recognition of predators will depend on relatively specific cues. We
might thus expect species to be less able to respond to the sounds of
evolutionarily and ontogenetically novel predators than to those of
historically important ones.
In summary, selection for optimal hunting will produce convergent morphology, while the constraints of effective communication will produce divergent sounds. We might therefore expect species to respond to the sights, but not the sounds, of evolutionarily novel predators.
In Australia, many small-bodied marsupials face extinction from predation
by red foxes introduced by Europeans in the last century
(Coman, 1995). Initially, this
detrimental impact of a novel predator may seem surprising because Australia
historically had a rich community of marsupial carnivores, including lions
(Thylacoleonidae), wolves (Thylacinidae), cats, and hyenas (Dasyuridae)
(Archer, 1981;
Robertshaw and Harden, 1989;
Wroe, 1999). Today many
species or populations survive only on predator-free offshore islands
(Johnson et al., 1989). We
studied one of these isolated populations, tammar wallabies (Macropus
eugenii) from Kangaroo Island, South Australia.
Tammars are a mid-sized (4-7 kg), moderately social wallabies found in
several geographically isolated and genetically distinct populations with
different histories of exposure to mammalian predators
(Cooper and McKenzie, 1997;
Croft, 1989;
Smith and Hinds, 1995).
Although they evolved with marsupial predators that preyed on tammar-sized
wallabies (e.g., Jones and Stoddard,
1998), tammars on Kangaroo Island have not typically been exposed
to mammalian predators since the island's isolation from the mainland about
9500 years ago (Lampert,
1979). The only substantial mammalian predator on Kangaroo Island,
the Tasmanian devil (Sarcophilus harrisii), is known only from the
fossil record (Pledge, 1979).
Devils are not effective predators on healthy, wallabysized animals
(Buchmann and Guiler 1977;
Jones, 1995; but see
Jones, 1994). There are no
records of the now extinct thylacine (Thylacinus cynocephalus),
dingoes (Canis lupus dingo), or red foxes (Vulpes vulpes)
having ever lived on the island. A small, low-density population of feral cats
(Felis catus) introduced in the last century, as well as domestic
farm dogs (Canis familiaris), are found around human settlements, the
majority of which are restricted to the eastern part of the island. A diurnal
raptor, the wedge-tailed eagle (Aquila audax), is a current, and
presumably historical, resident of Kangaroo Island that may prey on tammars.
Kangaroo Island tammars are thus ideally suited for study of the responses to
both evolutionarily novel and historically important predators.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Subjects and husbandry
We selected 19 adult tammar wallabies (16 females, 3 males) from a breeding colony maintained at the Macquarie University Fauna Park, Sydney, Australia. When not participating in the experiment, the wallabies lived in female-biased, mixed-sex aggregations. Subjects were either wild caught or captive bred from Kangaroo Island stock. Wild-caught subjects had lived in the Fauna Park for >2 years. For this study, we temporarily isolated each experimental subject in one of four 4 x 12 m fenced enclosures. The fences of each enclosure were covered with a 2 m high band of an opaque, black, woven synthetic fabric (weedmat) with an observation window cut into one of the small ends. Subjects were thus visually isolated from most movements outside the enclosure, but they could see the head and shoulders of the observer. The experimental enclosures contained shelter (arranged so as not to interfere with our vision) and ad libitum water. We provided food in a central location two times per day. An opening on each side of the enclosure revealed a stage on which a visual stimulus appeared, or beside which a speaker was hidden.
Stimulus conditions and justification
We selected a range of predator stimuli to include both a prototypical
historical predator and evolutionarily novel introduced predators. Wallaby
responses were compared to those evoked by control nonpredator stimuli and
blank trials in which no stimulus was presented. In addition, we studied the
response to a putative acoustic antipredator signal.
We exposed tammars to six visual treatments. Three predatory stimuli were chosen based on the pattern of evolutionary and ontogenetic isolation. A model thylacine, made of injection molded foam, was ontogenetically novel, but was a potentially important historical predator with which tammars should have had evolutionary experience. A cat was chosen because it was possibly ontogenetically familiar (feral and domestic cats sometimes wandered through the Fauna Park) but evolutionarily novel. A fox was both ontogenetically and evolutionarily novel. We also presented a taxidermically mounted wallaby to quantify the response to a nonpredatory, familiar vertebrate. We had two further controls. A cart control allowed us to quantify the response to the presentation device and its associated movement without a vertebrate stimulus, and a blank control allowed us to measure spontaneous behavioral change in the absence of a stimulus presentation.
We exposed tammars to five acoustic treatments. Tammar foot thumps were played back to document response to a putative antipredator signal. We used the calls from a wedgetailed eagle to represent acoustic cues from a historical and current predator. Similarly, the howls recorded from a pack of dingoes represent both an ontogenetically and evolutionarily novel predator-generated sound. We chose dingoes because they are more vocal than foxes. The remaining two conditions were controls. Playback of the song of an Australian magpie (Gymnorhina tibicen) allowed us to quantify the response to an ontogenetically and evolutionarily familiar non-predator that often sings from the ground. As in the visual treatments, a blank control allowed us to measure spontaneous behavioral change.
Experimental protocol
Subjects were removed from their normal enclosures and habituated to the
experimental setup for 3 days before the experiment began. Tammars were fed
half their daily allotment of food (kangaroo pellets and rolled oats) in the
morning (within 3 h of sunrise) and half in the late afternoon (within 3 h of
sunset). This provisioning schedule corresponded with the times at which
experiments were subsequently conducted. Tammars, although primarily
nocturnal, are active in the mornings and in the late afternoon
(Blumstein et al., 1999). In
nature, tammars begin to move from their day bed to communal foraging grounds
in the late afternoon and return to their day bed by the early morning. Thus,
these times approximate when tammars naturally move through their habitat and
would be most likely to encounter predators.
On the morning of the fourth day, and for the next six mornings, we exposed subjects to one of five visual stimuli or the blank control (Figure 1). Each stimulus was presented on a cart which ran along a wooden or aluminium track and appeared in the window, where it remained for 60 s before being pulled away.
|
On the afternoon of the fourth day and for the next five afternoons, we
exposed subjects to one of four acoustic stimuli or the blank control
(Figure 2). We used two
exemplars of each stimulus to sample natural variation in acoustic structure.
Dingo howls and wallaby thumps were recorded using a Sennheiser ME-67
microphone with a Sony TCD-5M cassette recorder. The wedge-tailed eagle
recordings were obtained from a commercial recording
(Buckingham and Jackson, 1985),
and the magpie recordings were obtained from a private collection. All sounds
were digitized (Digidesign Audiomedia II A/D board, Pro Tools software, sample
rate 44.1 kHz, 16-bit amplitude encoding), edited to create a 4-5 s acoustic
stimulus, and down-sampled to 8 bits for playback using a PowerBook 100
computer through a Sony SRS-77G powered speaker. Our playbacks were designed
to mimic the sound of an adjacent wallaby foot-thumping, a nearby eagle, a
more distant group of howling dingoes, and a nearby singing magpie. Thumps
were played back at an amplitude of 82 dB (A weighting; peak; ±1 dB
measured 1 m in front of the speaker); the other three stimuli were played
back at 92 dB (±1 dB).
|
Our goal was to present stimuli to subjects foraging in the center of the test enclosure to control for behavior, motivational state, and location. Pilot trials demonstrated that food type influenced vigilance patterns: wallabies eating only rolled oats had extended bouts of foraging with their nose held close to the ground. We climbed through the observation window and placed a handful of rolled oats at a predetermined central location 2-3 m from the window where the visual stimulus would appear or from the location of the hidden speaker. However, animals did not always feed when we set up a test trial. Some individuals rested or slept in the back of the enclosure, while others appeared to wait nervously for us to leave. Pilot trials demonstrated that wallabies that did not begin feeding within 15 min were not motivated to feed, so we waited a maximum of 30 min before presenting the stimulus. An additional criterion for visual stimuli was that subjects had to have an unobstructed view of the stage.
Stimuli were presented in a within-subjects repeated-measures design. Each subject was exposed to each stimulus once. To reduce the likelihood of habituation to our stimuli and to control for order effects, we tested enclosures in a random order, presented all stimuli and controls in a randomly generated but balanced order, and systematically varied the side of the test enclosure in which a stimulus appeared.
Data analysis
Subjects were videorecorded for 1 min before stimulus presentation, 1 min
during stimulus presentation, and for 3 min after stimulus presentation. A
single observer (J.C.D.) scored the videotapes to 0.1 s resolution using The
Observer 3.0 (Noldus Information
Technologies, 1995). We calculated the percentage of time
allocated to the three most common behaviors [foraging, looking, and
pentapedal locomotion (macropods use their tail when walking slowly; in this
context, we interpret such pentapedal locomotion as an exploratory behavior)].
Foraging was defined to include both bipedal and quadrupedal foraging.
Heightened looking was characterized by standing upright or rearing up on hind
legs. In all cases, we used the 1 min preceding stimulus presentation as a
baseline from which we calculated the difference in time allocation following
stimulus presentation. One way repeated-measures ANOVAs for each behavior
revealed no significant differences during the baseline period (visual
treatments: foraging, p =.60, looking, p =.65, pentapedal
locomotion, p =.15; acoustic treatments: foraging, p =.57;
looking, p =.95; pentapedal locomotion, p =.46).
Responses to the visual and acoustic stimuli were relatively transient. To quantify these changes in behavior, we examined time allocation in successive 15-s intervals, beginning at stimulus onset. For the visual treatments, we ignored the first 15 sec interval because all stimuli elicited a marked, but ephemeral, orienting response, and focused instead on the three 15-s intervals while the stimulus was present and stationary. We wanted to identify change in behavior due to the stimulus, independent of baseline rates, so we calculated the difference between baseline time allocation and that in each of the three 15-s intervals and modeled variation in time allocated to foraging, looking, and walking with a two-factor repeated-measures ANOVA. Some individuals also foot thumped in response to presentations of visual stimuli. We tested for a relationship between the probability of thumping and stimulus type using a Cochran's Q test, followed up by post-hoc pairwise McNemear's tests.
Responses to the 4 to 5-s acoustic stimulus presentation were very brief. We thus focused only on the 15-s interval that began with the playback. As with visual treatments, we calculated the difference between the percent baseline time allocation and that during the 15-s poststimulus interval and then fitted one-factor repeated-measures ANOVA models to explain variation in foraging, looking, and walking.
Statistical and power analyses were conducted using Statview 5.0
(SAS Institute, 1998),
SuperAnova 1.1 (Abacus Concepts Inc.,
1991), and G*Power
(Buchner et al., 1996).
Significance was set at p <.05. We visually examined residuals
from parametric models; these appeared to be normally distributed. For
repeated-measures ANOVAs, we report the Huynh-Feldt corrected p
values. For subsequent planned pairwise comparisons, we report Huynh-Feldt
corrected p values unadjusted for multiple comparisons
(Carmer and Swanson, 1973). We
also conducted post-hoc power analyses to estimate the sample size required to
differentiate key negative results with 70% power.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Visual treatments
Tammar wallabies responded to the presentation of all quadrupedal stimuli, particularly the cat and fox (Figure 3). Overall, foraging was inhibited, and both looking and locomotion were enhanced (foraging main effect: F5,90 = 3.87, p =.004; looking main effect: F5,90 = 3.03, p =.027; locomotion main effect: F5,90 = 3.32, p =.017). There was also a significant interaction between treatment and time for foraging (F10, 180 = 3.690, p =.0006) but not for the other behaviors (looking: F10,180 = 1.867, p =.074; locomotion: F10,180 =.879, p =.540). There were no significant main effects of time (largest F2, 36 = 1.58, p =.225).
|
We first compare the treatments by considering significant main effects. The fox significantly suppressed foraging relative to the cart (F1,90 = 9.871, p =.003) and blank (F1,90 = 13.416, p =.0005). Presentations of the fox also enhanced looking relative to the cart (F1,90 = 9.657, p =.006) and blank (F1,90 = 9.630, p =.006). Tammars responded similarly to the cat, which also suppressed foraging and enhanced looking (foraging: cat versus cart, F1,90 = 4.996, p =.029; cat versus blank, F1,90 = 7.596, p =.008; looking: cat versus cart, F1,90 = 4.719, p =.044; cat versus blank, F1,90 = 4.701, p =.044). Wallaby presentation also suppressed foraging (wallaby versus blank F1,90 = 5.994, p =.017). Locomotion too, was influenced by stimulus type. Tammars walked significantly more after seeing the fox and wallaby compared to the cart (fox, F1,90 = 8.070, p =.010; wallaby, F1,90 = 5.773, p =.027) and blank control (fox, F1,90 = 10.510, p =.004; wallaby, F1,90 = 7.861, p =.011).
We now consider the significant interaction between stimulus and time for foraging. The overall pattern was that subjects suppressed foraging most during fox and cat presentations (Figure 4). Subjects suppressed foraging significantly more after seeing the fox (F2,36 = 4.206, p =.028) and the cat (F2,36 = 4.486, p =.018) compared to the thylacine. Animals presented with no stimulus (the blank control) tended to forage less over time, illustrating an expected regression to the mean for this condition. There were significant stimulus x time interactions in all comparisons between the blank control and other treatments, which were each characterized by increased foraging over time (smallest F2,36 = 3.599, p <.038).
|
Post hoc power analyses emphasized the variation intrinsic in these response variables. For looking, we would have needed a median sample size of 148 subjects to better differentiate the nonsignificant responses to the stimuli (range = 30 for the wallaby-cart contrast to 388 for the wallaby—thylacine contrast). Similarly, for foraging, we would have needed a median sample size of 105 subjects to better differentiate the nonsignificant responses to the stimuli (range = 30 for the wallaby—cart contrast to 1420 for the wallaby—cat contrast). For walking, we would have needed a median sample size of 107 subjects to better differentiate the nonsignificant responses to the stimuli (range = 34 for the cart-thylacine contrast to 3350 for the cat—thylacine contrast).
The probability of tammars thumping varied significantly across our six
visual test conditions (Table
1; Cochran's Q = 18.8, p =.002). Subjects were
more likely to thump after seeing a fox than controls (McNemear's p
=.004 for blank, p =.022 for cart, p =.031 for wallaby).
However, a more conservative post hoc analysis using a sequential Bonferonni
test (Rice, 1989) suggests
that subjects were only significantly more likely to thump in response to the
fox than to the blank control.
|
Acoustic treatments
Wallabies responded to hearing sounds by foraging less and looking more
(Figure 5). This was reflected
by significant ANOVA main effects for foraging (F4,72 =
5.920, p =.0004) and looking (F4,72 = 5.309,
p =.0008), but not locomotion (F4,72 = 1.103,
p =.313).
|
Playback of thumps elicited significantly greater responses than both the blank control and the magpie control (foraging smaller F1,72 = 7.276, p =.009; looking smaller F1,72 = 6.230, p =.015). In contrast, the effects of predator sounds (dingoes and eagle) were not significantly different from those of the magpie song, but they each differed from the blank control (foraging: smaller F1,72 = 8.559, p =.005; looking: smaller F1,72 = 5.399, p =.023). The magpie control, however, did not significantly change behavior relative to the blank.
Post hoc power analyses again emphasized the level of variability in evoked responses. For looking, we would have needed a median sample size of 95 subjects (range = 27 for the magpie—eagle contrast to 292 for the dingo—magpie contrast) to better differentiate the nonsignificant responses to the different stimuli. Similarly, for foraging, we would have needed a median sample size of 85 subjects (range = 50 for the thumps—eagle contrast to 900 for the dingo—eagle contrast) to better differentiate nonsignificant responses.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The overall pattern of our results (Table 2) suggests that Kangaroo Island tammar wallabies respond to the sight of both evolutionarily and ontogenetically novel predators. The fox and cat stimuli inhibited foraging and enhanced vigilance relative to the cart and blank controls. Both the fox and the wallaby control stimulus increased pentapedal locomotion. Analysis of foot thumps, which we interpret as the highest level of antipredator behavior, revealed that the fox uniquely elicited more thumps than all three control stimuli. In contrast, the model thylacine, a species with which tammars shared a long evolutionary history (but for which a taxidermic mount was unavailable), did not elicit higher levels of response than the taxidermic mounts of novel predators.
|
The sounds of historical and novel predators did not modify tammar time
allocation, nor did they elicit foot thumps. However, tammars did respond to
playback of foot thumps by decreasing foraging and increasing vigilance
(Figure5). This response could
result from continued exposure to thumps which are elicited by people walking
through the Fauna Park. To our knowledge, this is the first demonstration that
the almost ubiquitous macropod foot thump functions as a conspecific
antipredator signal. Coulson
(1996) examined the situation
in which foot thumps were produced and concluded that for Western grey
kangaroos (Macropus fuliginosus), foot thumps may function to deter
pursuit from predators (Hasson,
1991). Alarm signals can have multiple targets and multiple
functions (Blumstein and Armitage,
1998; Evans 1997),
so these are not mutually exclusive interpretations. Thus, although tammars
were unresponsive to sounds produced by predators, they remain sensitive to
the alarm signals of conspecifics.
Taken together, these results suggest that Kangaroo Island tammar wallabies continue to respond to the sight of predators despite 9500 years of evolutionary isolation. These results are consistent with predictions about the generalizability of visual cues, in contrast with the diverse and species-specific nature of acoustic cues. Our study was designed to document the responses evoked by prototypical exemplars of historical and current predators. Future experiments would be required to identify the specific visual features (e.g., frontally positioned eyes, size, shape, body posture) that enabled tammars to respond to novel predators. Such experiments might also reveal why our thylacine model was deficient.
We suggest that recognition of visual features that are widely shared among
predators may be more likely to persist despite relaxed selection than
recognition of idiosyncratic acoustic cues. Such differences may also reflect
the relative importance of ontogeny, or very recent selection on Kangaroo
Island for responses to cats and dogs. It is possible, although unlikely, that
some of our subjects were exposed to cats and dogs before capture. Although
prey may be more likely initially to respond to visual cues from novel
predators than to their sounds, they still may be able to learn predator
vocalizations. Macedonia and Yount
(1991) found that a captive
colony of lemurs (Lemur catta) learned the calls of the
evolutionarily novel aerial predators present in their seminatural captive
environment. We know that tammars at the Macquarie Fauna Park heard
foot-thumps on a daily basis, but we do not have the detailed knowledge of
prior exposure to wedge-tailed eagle calls that would allow us to assess the
importance of learning for responsiveness to acoustic stimuli.
It is possible that olfactory cues associated with the taxidermic mounts explained some variation in the tammars' response. We are currently studying olfactory predator recognition. If olfactory cues were important, then we would expect also to find differences between treatments in time allocation before stimulus presentation, when the mounts were in position adjacent to the stage. However, there were no differences in baseline time allocation, suggesting that subjects did not respond to the smell of the fox or cat before they saw either stimulus.
There is substantial variation in the response of prey species to a period
of relaxed selection. Berger
(1998) found a decrease in
predator recognition abilities of North American ungulates isolated from
predators for only 50-100 years. In contrast, Byers
(1997) noted that the
antipredator behavior of pronghorn (Antilocapra americana) has
persisted, despite the extinction during the last ice age of the predators
with which they evolved. Similarly, Coss
(1999) found that ground
squirrels isolated from predators for 70,000-300,000 years have retained
predator recognition abilities.
What is it about isolation that influences the likelihood of a trait
persisting over time? Coss
(1999) suggested that
functionally integrated behaviors will likely persist over time. He argued
that, although there may be selection against a costly antipredator behavior
system, selection for components that are functional in other contexts may be
sufficient to maintain antipredator behavior. Avoidance responses involved in
antipredator behavior may have an underlying physiological response shared
with another system (e.g., social behavior). Thus, selection in the other
context may be sufficient to maintain antipredator behavior in the absence of
direct selection. Similarly, the continued presence of one class of predators
(e.g., raptors) may maintain functionally linked antipredator responses to
another class (e.g., carnivores; Blumstein and Daniel, unpublished data).
Finally, the cost of antipredator behavior must be considered. If responding
to the sight of a potential predator imposes no substantial energetic or
life-history costs, then a geographically isolated population will experience
no costly false positive identifications and may effectively have no cost to
maintaining some degree of responsiveness
(Magurran, 1999;
van Damme and Castilla,
1996).
Although tammars responded to the sight of a fox, which was an entirely
novel predator, changes in locomotor behavior were limited and baseline
behavior was resumed within 2 min after presentations. Many species of
macropods are endangered on mainland Australia, where their populations have
been decimated by introduced red foxes. There has been considerable recent
interest in prerelease training of endangered species to enhance their
responses to predators (Griffin et al., in
press; McLean et al.,
1994,
1999). Knowing how and when to
allocate scarce resources to prerelease predator training programs is vitally
important. We have suggested that training will be most successful for those
species that already possess rudimentary antipredator behavior
(Griffin et al., in press).
Results from the present study are encouraging because they demonstrate that
tammars retain the ability to respond to visual cues associated with
predators, despite a long period of isolation. We suggest that prerelease
predator training is likely to be successful for tammar wallabies, and
potentially for other endangered macropods, because even evolutionarily
isolated populations continue to respond to the sight of predators.
|
ACKNOWLEDGEMENTS |
|---|
We thank Des Cooper for ongoing advice about tammar wallabies and support of our research on tammars at the Fauna Park; Ron Claassens, Steve McLeod, Anne Mouland, and Scott Peel for wallaby wrangling; Des, John Rodger, and The Marsupial CRC for access to animals and facilities; Laura Lindow and Sydney's Taronga Zoo for letting us record dingoes; Walter Boles of the Australian Museum for help tracking down the wedgie recording; Sylvia Halkin for her magpie recording; Laura Anthony for help running experiments; Alan Taylor for statistical guidance; Linda Evans for comments on the manuscript; and Judy Brimage for her artistic efforts with our thylacine. Our macropod behavior research is supported by the Australian Cooperative Research Centre Programme. D.T.B. was supported by the Australian Research Council (Australian Postdoctoral Fellowship and small grant), Macquarie University (Macquarie University Research Grant and start-up funds). A.S.G. was supported by a MUIPGRA, the Australian Government Overseas Postgraduate Research Scholarship, and the Swiss Jangen-Pöhn Foundation. Research protocols were approved by the Macquarie University Animal Care and Ethics Committee (approval number 98023).
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Abacus Concepts, 1991. SuperAnova. Berkeley, California: Abacus Concepts.
Archer M, 1981. A review of the origins and radiations of Australian mammals. In: Ecological biogeography of Australia (Keast A, ed). The Hague: W. Junk; 1436-1488.
Berger J, 1998. Future prey: some consequences of the loss and restoration of large carnivores. In: Behavioral ecology and conservation biology (Caro TM, ed). New York: Oxford University Press; 80-100.
Berger J, 1999. Anthropogenic extinction of top carnivores and interspecific animal behaviour: implications of the rapid decoupling of a web involving wolves, bears, moose and ravens. Proc R Soc Lond B 266: 2261-2267.[Medline]
Blumstein DT, Armitage KB. 1998. Why do yellow-bellied marmots call? Anim Behav 56: 1053-1055.[Medline]
Blumstein DT, Evans CS, Daniel JC, 1999. An experimental study of behavioural group size effects in tammar wallabies (Macropus eugenii). Anim Behav 58: 351-360.[ISI][Medline]
Buchmann OLK, Guiler ER, 1977. Behaviour and ecology of the Tasmanian devil, Sarcophilus harrisii. In: The biology of marsupials (Stonehouse B, Gilmore D, eds). London: Macmillan; 155-168.
Buchner A, Fawl F, Erdfelder E, 1996. G*Power: A priori, post-hoc, and compromise power analysis for the Macintosh (version 2.1.1). Trier, Germany: University of Trier.
Buckingham R, Jackson L, 1985. A field guide to birdsong, cassette 2: Rufous night heron to chestnut rail. Nunawading, Victoria: Bird Observers Club.
Byers JA, 1997. American pronghorn: social adaptations and the ghosts of predators past. Chicago: University of Chicago Press.
Carmer SG, Swanson MR, 1973. An evaluation of ten pairwise multiple comparison procedures by Monte Carlo methods. J Am Stat Assoc 68: 66-74.
Coman BJ, 1995. Fox. In: The mammals of Australia (Strahan R, ed). Chatswood, New South Wales: Reed Books; 698-699.
Cooper DW, McKenzie LM, 1997. Genetics of tammar wallabies. In: Marsupial biology: recent research, new perspectives (Saunders NR, Hinds LA, ed). Sydney: University of New South Wales Press; 120-131.
Coss RG, 1991. Evolutionary persistence of memory-like processes. Concepts Neurosci 2: 129-168.
Coss RG, 1999. Effects of relaxed natural selection on the evolution of behavior. In: Geographic variation in behavior: perspectives on evolutionary mechanisms (Foster SA, Endler JA, eds). Oxford: Oxford University Press; 180-208.
Coss RG, Goldthwaite RO, 1995. The persistence of old designs for perception. Perspect Ethology 11: 83-148.
Coulson G, 1996. Anti-predator behaviour in marsupials. In: Comparison of marsupial and placental behaviour (Croft DB, Gansloßer U, eds). Fürth: Filander Verlag GmbH; 158-186.
Croft DB, 1989. Social organization of the Macropodoidea. In: Kangaroos, wallabies and rat-kangaroos, vol. 2 (Grigg G, Jarman P, Hume I, eds). Chipping Norton, New South Wales: Surrey Beatty & Sons; 505-525.
Curio E, 1966. How finches react to predators. Animals 9: 142-143.
Curio E, 1975. The functional organization of anti-predator behaviour in the pied flycatcher: a study of avian visual perception. Anim Behav 23: 1-115.[ISI][Medline]
Curio E, 1993. Proximate and developmental aspects of antipredator behavior. Adv Study Behav 22: 135-238.
Estes RD, 1991. The behavior guide to African mammals: including hoofed mammals, carnivores, primates. Berkeley: University of California Press.
Evans CS, 1997. Referential signalling. Perspect Ethol 12: 99-143.
Evans CS, Macedonia JM, Marler P, 1993. Effects of apparent size and speed on the response of chickens (Gallus gallus) to computer-generated simulations of aerial predators. Anim Behav 46: 1-11.
Griffin AS, Blumstein DT, Evans CS, in press. Training animals about predators: a critical review and synthesis. Conserv Biol.
Hasson O, 1991. Pursuit-deterrent signals: communication between prey and predators. Trends Ecol Evol 6: 325-329.
Hauser MD, Wrangham RW, 1990. Recognition of predator and competitor calls in non-human primates and birds: a preliminary report. Ethology 86: 116-130.
Hendrie CA, Weiss SM, Eilam D, 1998. Behavioural response of wild rodents to the calls of an owl: a comparative study. J Zool 245: 439-446.
Hirsch SM, Bolles RC, 1980. On the ability of prey to recognize predators. Z Tierpsychol 54: 71-84.
Johnson KA, Burbidge AA, McKenzie NL, 1989. Australian Macropodoidea: status, causes of decline and future research and management. In: Kangaroos, wallabies and rat-kangaroos, vol. 2 (Grigg G, Jarman P, Hume I, eds). Chipping Norton, New South Wales: Surrey Beatty & Sons; 641-657.
Jones M, 1995. Tasmanian devil. In: The mammals of Australia (Strahan R, ed). Chatswood, New South Wales: Reed Books; 82-84.
Jones ME, 1994. Dining with the devil. Aust Nat Hist 24: 32-41.
Jones ME, Stoddard DM, 1998. Reconstruction of the predator behaviour of the extinct marsupial thylacine (Thylacinus cynocehalus). J Zool 246: 239-246.
Kavailers M, 1990. Responsiveness of deer mice to a predator, the short-tailed weasel: population differences and neuromodulatory mechanisms. Physiol Zool 63: 388-407.
Lampert RJ, 1979. Aborigines. In: Natural history of Kangaroo Island (Tyler MJ, Twidale CR, Ling JK, eds). Adelaide: Royal Society of South Australia; 81-89.
Levesley PB, Magurran AE, 1988. Population differences in the reaction of minnows to alarm substance. J Fish Biol 32: 699-706.
Macedonia JM, Yount PL, 1991. Auditory assessment of avian predator threat in semi-captive ringtailed lemurs (Lemur catta). Primates 32: 169-182.
Magurran AE, 1999. The causes and consequences of geographic variation in antipredator behavior: perspectives from fish populations. In: Geographic variation in behavior: perspectives on evolutionary mechanisms (Foster SA, Endler JA, ed.). New York: Oxford University Press; 139-163.
McLean IG, Hölzer C, Studholme BJS, 1999. Teaching predator-recognition to a naive bird: implications for management. Biol Cons 87: 123-130.
McLean IG, Lundie-Jenkins G, Jarman PJ, 1994. Training captive rufous hare-wallabies to recognize predators. In: Reintroduction biology of Australian and New Zealand fauna (Serena M, ed). Chipping Norton, New South Wales: Surrey Beatty & Sons; 177-182.
Noë R, Bshary R, 1997. The formation of red colobus-diana monkey associations under predation pressure from chimpanzees. Proc R Soc Lond B 264: 253-259.[Medline]
Noldus Information Technologies, 1995. The observer 3.0. Wageningen, The Netherlands: Noldus Information Technologies.
Pledge NS, 1979. Fossil vertebrates. In: Natural history of Kangaroo Island (Tyler MJ, Twidale CR, Ling JK, eds). Adelaide: Royal Society of South Australia; 123-127.
Pressley PH, 1981. Parental effort and the evolution of nest-guarding tactics in the threespine stickleback, Gasterosteus aculeatus L. Evolution 35: 282-295.
Rice WR, 1989. Analyzing tables of statistical tests. Evolution 43: 223-225.[ISI]
Robertshaw JD, Harden RH, 1989. Predation on Macropodoidea: a review. In: Kangaroos, wallabies and rat-kangaroos, vol. 2 (Grigg G, Jarman P, Hume I, eds). Chipping Norton, New South Wales: Surrey Beatty & Sons; 735-753.
SAS Institute, 1998. Statview 5.0. Cary, North Carolina: SAS Institute.
Schaller GB, 1972. The Serengeti lion: a study of predator-prey relations. Chicago: University of Chicago Press.
Seghers BH, 1974. Schooling behavior in the guppy (Poecilia reticulata): an evolutionary response to predation. Evolution 28: 486-489.[ISI]
Smith MJ, Hinds L, 1995. Tammar wallaby. In: The mammals of Australia (Strahan R, ed). Chatswood, New South Wales: Reed Books; 329-331.
Swaisgood RR, Rowe MP, Owings DH, 1999. Assessment of rattlesnake dangerousness by California ground squirrels: exploitation of cues from rattling sounds. Anim Behav 57: 1301-1310.[ISI][Medline]
Tinbergen N, 1951. The study of instinct. Oxford: Oxford University Press.
Van Damme R, Castilla AM, 1996. Chemosensory predator recognition in the lizard Podarcis hispanica: effects of predation pressure relaxation. J Chem Ecol 22: 13-22.
Wroe S, 1999. Killer kangaroos. Sci Am 280(5): 68-74.[ISI][Medline]
Zuberbühler K, Noë R, Seyfarth RM, 1997. Diana monkey long-distance calls: messages for conspecifics and predators. Anim Behav 53: 589-604.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  B. W. Brook and D. M. J. S. Bowman Explaining the Pleistocene megafaunal extinctions: Models, chronologies, and assumptions PNAS, November 12, 2002; 99(23): 14624 - 14627. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. Randall Evolution and Function of Drumming as Communication in Mammals Integr. Comp. Biol., October 1, 2001; 41(5): 1143 - 1156. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||