Behavioral Ecology Vol. 11 No. 1: 102-109
© 2000 International Society for Behavioral Ecology
Trade-off in short- and long-distance communication in túngara (Physalaemus pustulosus) and cricket (Acris crepitans) frogs
a Department of Zoology, University of Texas, Austin, TX 78712, USA b Department of Biological Sciences, Central Washington University, Ellensburg, WA 98926, USA c Department of Psychology, University of Texas, Austin, TX 78712, USA d Smithsonian Tropical Research Institute, Apd. 2072 Balboa, Panama
Address correspondence to L. Sun, Department of Biological Sciences, Central Washington University, Ellensburg, WA 98926, USA. E-mail: lixing{at}cwu.edu .
Received 5 May 1998; accepted 23 June 1999.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Female phonotaxis in túngara (Physalaemus pustulosus) and cricket (Acris crepitans) frogs is biased toward male advertisement calls or call components of lower frequency. This behavioral bias might result in part from a mismatch between the spectral characteristics of the advertisement call and the most sensitive frequency of the peripheral end organ implicated in reception of these sounds. In both species, females are tuned to frequencies lower than average for the calls in their population. This mismatch, however, represents the situation during short-distance communication. Female frogs can also use the call to detect choruses at long distances, and the spectral distribution of call energy can vary with transmission distance. We used computer simulations to test the hypothesis that there is a better match between tuning and call spectral energy at long distances from the calling male than at short distances by comparing the performance (sound energy received) of the natural tuning curve relative to an optimal tuning curve (i.e., one centered at the call's dominant frequency). The relative performance of the natural tuning curve increased with distance in túngara frogs. For the two subspecies of cricket frogs, however, the relative performance decreased at longer distances. The performance did not equal the optimal tuning curve at the distances tested. The results indicate that the relationship between calls and auditory tuning cannot be optimal for both long and short distance reception. The relationship between female tuning and call dominant frequency may represent a compromise between short and long distance communication, and the bias toward short or long distances may vary among species.
Key words: Acris crepitans, auditory system, communication, cricket frogs, Physalaemus pustulosus, sensory biases, sexual selection, túngara frogs.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Female mating preferences are an important selection force influencing the evolution of male secondary sexual characters and are often influenced by male display traits that function as mating signals (Andersson, 1994
;
Darwin, 1871). Female
preferences for different variants of these signals result from an interaction
of the external signal with a set of internal physiological processes
including neural, cognitive, and hormonal functions. Although these
physiological systems mediate female mate choice at close distance, they serve
a variety of other functions as well. Selection acting on these internal
processes outside of the context of short-distance mate assessment might
generate biases in female mating preferences as unintended consequences or
pleiotropic effects; these biases, in turn, generate sexual selection on male
signals (e.g., Christy, 1995;
Dawkins and Guilford, 1996;
Eberhard, 1993;
Endler, 1992;
Guilford and Dawkins, 1993;
Kirkpatrick and Ryan, 1991;
Ryan, 1990,
1997;
Shaw, 1995;
West-Eberhard, 1979). For
example, signals might function as beacons for females to locate distant
aggregates of calling males (i.e., choruses), as well as signals that females
assess when choosing individual mates at short distances. In this study we
tested the hypothesis that selection acting on aspects of the signal and
receiver for long-distance communication might have unintended consequences on
short-distance communication that generates sexual selection.
The courtship signals that males use to attract females for mating have
played central roles in studies of speciation
(Blair, 1958;
Mayr, 1963) and sexual
selection (Andersson, 1994;
Darwin, 1871). Anurans have
been a useful model system in these studies due to the relative stereotypy of
male advertisement calls, the ability to assess female preferences using
phonotaxis, and the potential to uncover the sensory bases for behavioral
biases through neurophysiological studies of the auditory system
(Capranica, 1976; Gerhardt,
1994a,
b;
Ryan, 1991). In all phonotaxis
studies to date, females have shown preferences for a conspecific call versus
a heterospecific call when they are presented with such a choice in a
two-speaker, simultaneous-choice paradigm
(Gerhardt, 1994b;
Littlejohn and Michaud, 1959;
Rand, 1988).
Call detection and recognition involve reception of the call by peripheral
end organs in the inner ear for processing through a series of auditory
centers in the central nervous system
(Feng et al., 1990;
Fritzsch et al., 1988;
Hall, 1994;
Wilczynski and Capranica,
1984). Two peripheral end organs can be involved in the initial
reception of the call: the amphibian papilla (AP) and the basilar papilla
(BP). These end organs differ in a variety of ways, including their frequency
sensitivities. Within a species, the AP is always more sensitive to lower
frequencies than is the BP. For most species, the AP is most sensitive to
frequencies below 1500 Hz and the BP to frequencies above 1500 Hz
(Wilczynski and Capranica,
1984; Zakon and Wilczynski,
1988; Fox, 1995).
In some species the peak sensitivities of both the AP and the BP match
concentrations of spectral energy in the call, whereas in other species most
of the call energy lies within the most sensitive frequency range of either
the AP or the BP. The match between the tuning of the peripheral end organs
and the calls' dominant frequencies is thought to be one of the
characteristics of the communication system that facilitates female
preferences for conspecific calls and for the calls of certain males within
the species (e.g., Ryan,
1994).
The systems
In the túngara frog, Physalaemus
pustulosus, males produce a frequency-modulated call component, the
whine, which consists of a fundamental frequency with substantial energy in
the first four harmonics (Figure
1). In a typical call, the dominant frequency of the whine is 700
Hz, and most of the call energy is in the fundamental frequency, which sweeps
from about 900 to 400 Hz in 300 ms. The whine can be produced alone, or it can
be followed by additional components, chucks. A typical chuck is 35 ms in
duration and has a fundamental frequency of 225 Hz with 14 harmonics that
contain substantial energy. More than 90% of the chuck's energy lies above
1500 Hz, and the average dominant frequency of the chuck is 2500 Hz
(Ryan et al., 1990b). Although
females show phonotactic responses to the simple whine, they prefer whines
with chucks (Rand and Ryan,
1981).
|
The most sensitive frequency of the AP in
túngara frogs is about 700 Hz, closely
matching the dominant frequency of the whine
(Ryan et al., 1990b), and
phonotaxis experiments have shown that the portions of the fundamental
frequency of the whine with energy close to 700 Hz are critical for call
recognition (Rand et al.,
1992; Wilczynski et al.,
1995). The most sensitive frequency of the BP is 2200 Hz, which is
in the range of most of the spectral energy in the chuck
(Ryan et al., 1990b), although
this tuning is below the average dominant frequency of the chuck in the
population we studied (2500 Hz). Phonotaxis experiments show that the higher
harmonics in the chuck (1600-2800 Hz) will increase call attractiveness, but
this is not true for the lower harmonics (200-1400 Hz) at their normal
amplitudes (Rand et al., 1992;
Wilczynski et al., 1995).
Female túngara frogs tend to prefer calls
with dominant frequencies in the lower end of the population's range to those
at the higher end (Ryan, 1980,
1983;
Wilczynski et al., 1995). We
previously hypothesized that females prefer lowerfrequency chucks because this
effects a better match between call and BP tuning and thus increases the
amount of neural stimulation of this peripheral end organ. Computer models
that integrated natural chucks with the tuning function of an average BP were
consistent with this hypothesis (Ryan et
al., 1990b).
Cricket frogs, Acris crepitans, produce rapidly pulsed calls with
a dominant frequency of about 3500 Hz, depending on the population studied
(Figure 1;
Capranica et al., 1973;
Nevo and Capranica, 1985;
Ryan and Wilczynski, 1988;
Wilczynski and Ryan, 1999).
All of the call's energy falls within the range of the most sensitive
frequencies of the BP, with little or no call energy below 1500 Hz
(Capranica et al., 1973;
Keddy-Hector et al., 1992;
Wilczynski et al., 1992).
Although call frequency and BP tuning tend to covary among populations,
females are usually tuned below the average call dominant frequency in their
population. When temporal properties of the calls are controlled, females
often prefer calls with lower dominant frequencies
(Ryan and Wilczynski, 1988;
Ryan et al., 1992). Therefore,
the mismatch of BP tuning and average call dominant frequency is in part
responsible for the low frequency preference exhibited by females
(Ryan et al., 1992), although
the response of peripheral end organs may not be a good predictor of the
behavior ultimately exhibited in nature
(Schwartz and Gerhardt,
1998).
The question
The slight mismatch between BP tuning and call dominant frequency could be
explained by several hypotheses. One hypothesis is based on the fact that BP
tuning can vary with body size within a species, as in cricket frogs
(Keddy-Hector et al., 1992).
Because females are larger than males in most species of frogs
(Shine, 1979), females will
tend to have a BP tuned to lower frequencies than the males' BP
(McClelland et al., 1997;
Wilczynski et al., 1984;
Zakon and Wilczynski, 1988).
If selection favors a male to be tuned to calls of other males, then, by
default, females will be tuned lower than the male call. This might be a
plausible explanation, as in some species the perceived intensity of the
nearest neighbors' calls is thought to mediate nearest neighbor distances
(Brenowitz et al., 1984;
Wilczynski, 1986). A second
hypothesis views this mismatch as an adaptation for choosing high-quality
males. Larger males usually produce lower frequency calls; if females derived
an advantage from mating with large males, selection might favor females being
attracted to the lower frequency calls of larger males. In some frogs females
gain increased fertilization success from mating with larger males
(Bourne, 1993;
Robertson, 1990;
Ryan, 1985), but there is no
evidence that such mating preferences increase the genetic quality of the
offspring (Howard et al.,
1994). Thus, other hypotheses are also likely.
Unlike many other animals, such as some insects, most frogs do not have
different signals that are used in short- versus long-distance communication.
The male's advertisement call typically acts as a beacon to attract females
from long distances and as a courtship signal at short distances, but long and
short-distance communication can act under different constraints. It is well
known that sound degradation tends to be frequency dependent
(Dusenbery, 1992). In general,
higher frequencies degrade more rapidly due in part to the greater
susceptibility to the scattering of shorter wavelengths
(Forrest, 1994;
Wiley and Richards, 1978).
Therefore, lower frequencies lose less energy than the higher frequencies in
the signal and so become relatively more prominent in long-distance
communication. If the relative spectral distribution of call energy shifts to
lower frequencies with distance from the source, then it would not be possible
to optimize the match between tuning and call frequency at both long and short
distances. Therefore, we tested a third hypothesis that the mismatch between
BP tuning and call dominant frequency in short-distance communication is a
trade-off for a better match between signal and receiver to facilitate
long-distance communication.
This hypothesis makes two predictions. The first prediction is that there is a better match between tuning and call frequency at long-distance interactions, but this necessitates a greater mismatch at short distances. If supported this suggests that the relationship between signal and receiver cannot be equally efficient at all distances, and the relationship thus results from compromises between short- and long-distance communication. The second prediction is that the match between tuning and call frequency is not only better at long than short distances, but that it is optimal. This would suggest that selection has maximized the signal—receiver match for long-distance communication despite its reduced efficiency in short-distance communication. This hypothesis might be especially appealing because, at its most sensitive frequencies, the BP has higher neural thresholds than the AP, i.e., it is less sensitive to sound. Because signals attenuate with distance, improving the match should improve sensitivity to low-amplitude calls. Thus optimizing the match between tuning and call frequency when the BP is involved might be more important for long-distance communication than for short-distance communication.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
General approach
We tested the hypothesis that the apparent mismatch between tuning and call frequency at short distances is a consequence of a better match between tuning and call frequency at long distances (Figure 2). We did this by constructing computer models of BP tuning based on the audiograms obtained in previous studies (Ryan et al., 1990
; Wilczynski et al.,
1992), then used these models as filters through which we passed
digitized versions of natural calls tape recorded at different distances from
a calling frog or a speaker broadcasting a call (see below). Two pairs of
models were constructed for each taxon investigated. One filter model in the
pair exhibited the tuning characteristics of the average female BP and thus
was tuned to a frequency below the male call's average dominant frequency; we
refer to this as the "natural filter." The second filter was
similar to the first but was adjusted so that its peak tuning (or best
excitatory frequency, BEF) matched the call's dominant frequency; we refer to
this as the "optimal filter." We evaluated the performance of each
filter by determining the amount of energy in the call after it was passed
through each filter. We then calculated the relative performance, which is the
ratio of the performance of the natural filter and the optimal filter.
Finally, we used these data to test the prediction that the relative
performance of the natural filter increased with distance from the source or,
more specifically, that the difference in relative performance (DRP) at long
and short distances was significantly different from zero.
|
), will change as a result of
frequency-dependent attenuation. The performance of the natural filter
(RMSn), which models the filtering properties of the frog's basilar
papilla, is the amount of energy in the call after it is passed through this
filter. Note that the peak sensitivity of the filter does not match the call's
dominant frequency. The hypothetical optimal filter, however, has its peak
sensitivity centered on the call's dominant frequency. Its performance
(RMSo) is also determined by measuring the amount of energy in the
call after it is passed through this filter. The performance of the natural
filter is compared with the optimal filter to determine the natural filter's
relative performance. This is done for both the short- and long-distance
interaction.
Degraded calls
Physalaemus pustulosus
We used calls of túngara frogs from a
previous study of call degradation and attenuation
(Ryan, 1986). Calls of male
túngara frogs were recorded in a large, open
field at ground level (the height of the sender and receiver in this
communication system) in Gamboa, Republic of Panama. Two microphones
(Sennheiser ME 80 with KU-3 power modules) were placed 1 m and 11.6 m away
from calling frogs, and we recorded calls from at least three individuals on
two channels of a stereo Sony TCD-5M tape recorder with metal tape. The
frequency response of the recording system was flat (within 1 dB) for the
frequency range (500-6000 Hz) of the frog calls (see
Ryan, 1986, for more details).
The spectra of all calls were close to the average of the species.
Acris crepitans
We used calls of cricket frogs from a previous study of call degradation
and attenuation (Ryan et al.,
1990a). The calls were from six male cricket frogs (8-11
calls/frog) each from a population of Acris crepitans crepitans in
Polk County, east Texas, USA, pine forest, which is characterized by
relatively dense vegetation, and a population of A. c. blanchardi at
Gill Ranch in Travis County, central Texas, USA, which is in juniper—oak
woodland, a generally open habitat (see
Ryan and Wilczynski, 1991, for
details of geographic variation in calls). Because our purpose was to
determine the influence of environmental factors on the performance of frog
calls, the experimental unit was an individual call, rather than an individual
frog. We broadcast the calls with a Sony TCD-5M tape recorder and a small
extension speaker (Mineroff) at about 100 dB sound pressure level (SPL) (20
µPa) at 1 m from the source, which is within the range of the natural
calling intensity of cricket frogs. We recorded calls simultaneously with
stereo tape recorders (Marantz PMD 420) and two microphones (Sennheiser ME 80
with K3U power modules) placed at 1 m and 16 m from the source. An earlier
study that extrapolated from recordings of neural thresholds and call
attenuation suggested that females could detect a call of the same SPL from 18
m (Fox, 1988), although
whether females use these calls to find choruses over this distance is
unknown.
All experiments were conducted in early evening, the time when cricket frogs usually begin their calling. Degradation of the A. c. blanchardi subspecies calls was measured in the open habitat at Gill Ranch in Austin, Texas, which is typical of this subspecies' environment. Degradation of calls of the A. c. crepitans subspecies was measured in a closed habitat of dense pine forest at Stengl Ranch in Bastrop County, Texas, which is the habitat type of this subspecies. To determine if any results obtained were a necessary consequence of frog calls and auditory systems independent of habitat characteristics, we also measured degradation for each subspecies call in the non-native habitat.
Basilar papilla tuning
Details about neurophysiological recordings of the response properties of
the BP and the audiograms for túngara frogs
from Panama and the two cricket frog subspecies from two populations in Texas
are available elsewhere (Ryan et al.,
1990a,
b,
1992;
Wilczynski et al., 1992). Data
for the túngara frogs were obtained by
recording acoustically evoked multiunit activity from the midbrain. Data for
the cricket frogs were obtained by recording acoustically evoked activity in
individual VIIIth nerve afferent fibers. Data obtained from the two sources
may provide different estimates of absolute threshold but generally agree well
on basic tuning properties such as the best excitatory frequency (BEF)
(Fox, 1995).
For each taxon, an average BP tuning function was first estimated. The
audiograms for male and female túngara frogs
do not differ, and data from both sexes were used to determine the average
audiogram (Ryan et al., 1990).
BP tuning differed between the sexes in cricket frogs, thus only data from
females were used. First, the average BEF (frequency with the lowest
threshold) was calculated for each species and subspecies. Second, the average
band widths of the tuning at 10 dB above threshold and at 20 dB above
threshold were determined. Third, the asymmetry of natural tuning curves, for
which the high-frequency slope is steeper than the low-frequency slope, was
factored into the model by calculating the average frequency points on the
high and low flanks of the physiologically recorded tuning curves at 10 and 20
dB above threshold, determining the relative displacements of each above and
below the BEF, and adjusting the average curve accordingly while keeping the
bandwidth at the average value. The resultant average BP tuning curve then
served as a filter function, which we termed the "natural" tuning
function of each of the three groups (the
túngara frog and the two subspecies of cricket
frogs, see Figure 2).
We obtained an optimal tuning curve for each of the three groups by
shifting the natural tuning curve by a certain amount (by adding a constant:
420 Hz for P. pustulosus, 740 Hz for A. crepitans
blanchardi, and 80 Hz for A. c. crepitans) such that the BEF of
the shifted curve matched the dominant frequency of the average male call for
that group. The natural and optimal tuning curves were then converted into
fast Fourier transformation (FFT) files in the sound analysis-synthesis
program Signal (Engineering Design; Beeman,
1994), where they could be used to filter the calls by convolving
the filter function with the call spectrum.
Data analysis
We used the Signal-RTS software to digitize the recorded calls (sampling
rate: 12500 Hz for the túngara frogs, and
25000 Hz for the cricket frogs, higher than 1.5 times of the Nyquist frequency
for all calls recorded). We analyzed the chuck of the
túngara frog call and the entire call for the
cricket frogs (see above).
Before analysis, calls were digitally band-pass filtered within Signal at
frequencies of 500-4000 Hz for túngara frog
calls and 1500-6000 Hz for cricket frog calls to minimize the effect that
background noise had on the calls' spectral distribution. We then selected the
same calls recorded from the short and long distances. We calculated the FFT
size (1024 for the túngara frog and 2048 for
the cricket frog) and compared the power spectra for each signal recorded at
the two distances. The power spectra of short- and long-distance calls were
then adjusted to the same average call energy, as measured by the root mean
square (RMS) without equalizing energy level. We filtered the same calls
recorded from the two distances with the two BP tuning functions, the natural
and optimal filters. We calculated the relative performance (RP) of the
natural and optimal filter for the same calls as:
 | (1) |
 | (2) |
For túngara frogs, we tested the DRP using a Z test (two tailed) against the null hypothesis of DRP = 0, which indicates neither improvement nor deterioration in the performances between short- and long-distance communication. For cricket frogs, we used a factorial analysis of variance with habitat and subspecies as the 2 factors and 60 replications. If the interactions between the two main effects were significant, we used a t test (two tailed) to examine all simple effects. All tests were conducted after we had found that the normality assumption was not violated using the original data. The significance level for all tests was.05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
The DRP between the optimal and natural filters for the túngara frog was 11.68% ± 9.50% (SD) between the short- and long-distance calls (z = 13.70, df = 123, p <.001; Figure 3). Thus the performance of the natural filter, when compared with that of the optimal filter, was greater at the longer distance than it was at the shorter distance. These results are consistent with the prediction that the tuning-call mismatch in short-distance communication is related to a better match in long-distance communication. For the long distance, the natural tuning curves consistently underperformed that of the optimal tuning curve (on average 0.47% worse in the túngara frog and 0.18% in the cricket frog, respectively). These data did not support the hypothesis that the natural tuning curve is optimized for maximizing long-distance communication.
|
The DRP was -2.39% ± 2.25% (SD) for A. c. crepitans and -7.21% ± 18.09% (SD) for A. c. blanchardi in their respective native habitat. In the non-native habitat, the DRP was -3.79% ± 1.72% (SD) for A. c. crepitans and 14.60% ± 11.33% (SD) for A. c. blanchardi. An analysis of variance showed that the main effects of habitat and subspecies were both significant (F1,236 = 177.32, p <.001 for habitat and F1,236 = 78.43, p <.0001 for subspecies), but the interaction between habitat and subspecies was also significant (F1,236 = 229.19, p <.001). We thus investigated the simple effect for the habitat factor. For both subspecies of cricket frogs, the RP increased with distance in their respective habitats (A. c. blanchardi: t = 6.31, df = 59, p <.001; A. c. crepitans: t = 20.12, df = 59, p <.001; Figure 4). Thus, for both subspecies in their native habitat, the performance of the natural filter was worse at the long distance than at the short distance. At neither distance did the natural filter perform as well as the optimal filter. In the non-native habitat, RP increased with distance for A. c. crepitans (t = 10.82, df = 59, p <.001), but decreased with distance for A. c. blanchardi (t = 14.90, df = 59, p <.001).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
We tested the hypothesis that the mismatch between BP tuning and call dominant frequency in túngara frogs and cricket frogs in short-distance communication is an epiphenomenon of a better match between signal and receiver to facilitate long-distance communication. We tested two predictions from this hypothesis: the relative performance of the tuning curves is better for long-distance than for short-distance communication, and the match between BP tuning and call frequency is optimal for long-distance communication. Our results from túngara frogs support the first prediction: the performance of the natural tuning curve relative to the optimal one increased with distance. In contrast, the relative performance of the natural filter in cricket frogs decreased as distance increased for both subspecies. These results are the opposite of the first prediction. In our analysis of both túngara frogs and cricket frogs, our results fail to support the more extreme optimization prediction because at both short and long distances the natural tuning curves performed worse than the theoretical optimum.
If the performance of the natural tuning curve relative to the optimal
filter continued to increase with distance from the source, then at some
distance the natural tuning curves' performance would equal that of the
optimal for the túngara frog. We used only one
estimate of a long distance in this study for each species, 11.6 m from the
source for túngara frogs and 16 m for cricket
frogs. Assume that incremental improvement in the performance of the natural
filter is linear for simplicity in calculation. Extrapolating from the rate at
which the relative performance of the natural tuning curve increases, the
natural tuning curve would perform as well as the optimal tuning curve at
about 94 m from the source for túngara frogs.
This is far beyond the estimates of threshold for response derived from
neurophysiological data (Fox,
1988).
Our results lead us to conclude that selection for long-distance
communication in certain circumstances might contribute to the mismatch
between tuning and call frequency we observe in short-distance communication,
but there is no evidence to support the notion that the signal—receiver
match is optimized for long-distance communication. This lack of optimization
might have several causes. We doubt that there are any nonselection
constraints, in the sense of those discussed by Maynard Smith et al.
(1985), that would prohibit
the female's BP from being tuned to higher frequencies, as many small frogs
are tuned to much higher frequencies than are
túngara frogs (e.g.,
Wilczynski et al., 1993). But
achieving an optimal match between tuning and call frequency for long-distance
communication would result in an even larger mismatch between tuning and call
frequency in short-distance communication than we already observe. This might
be selectively disadvantageous for short-distance communication despite the
fact that the absolute stimulus energy available to receivers is larger at
short distances than long distances. The observed relationship between tuning
and call frequency might have resulted from a compromise of contrasting
selection forces to simultaneously increase efficiency for both short- and
long-distance communication.
The notion that the relationship between peak call energy and tuning of the auditory system is a compromise between long- and short-distance communication is, in the most general sense, consistent with the results from both the túngara frog and both subspecies of the cricket frog. In no case was the natural filtering function of the auditory system as good as the theoretical optimum, suggesting that the relationship between the spectral composition of the call and the tuning properties of the auditory system is not optimal for communication over any distance (at least none of the distances measured here, which were all within the range over which frogs interact). For each species, however, the signal—receiver system was better at one distance than at another. For the túngara frog, the system performed better at long-distance communication, whereas for both cricket frog subspecies, the system performed better at short-distance communication.
At present, we cannot address the reason that the species adopt different
strategies, as they differ in too many factors. They are phylogenetically
distinct, and one nonadaptive explanation would be the retention of an
ancestral character in either the call or auditory properties that bias the
responses. For Physalaemus at least, phylogeny does influence call
recognition and preference (Ryan and Rand,
1995; Ryan et al.,
1990). The species in this genus breed in different habitats, and
the results testing the signal—receiver interaction of cricket frogs in
non-native habitats shows that habitat and vegetation characteristics can play
a pivotal role in the transmission and reception of sound measured by the
performance of natural filters. In fact, the results from A. c.
blanchardi in the non-native habitat demonstrate empirically that it is
physically possible for cricket frog calls and tuning to be better matched at
long distances than at short distances, but that the opposite specialization
actually characterized the system in the habitat in which the calls are used.
The behavioral ecology is different in the two species, with
túngara frogs often calling from small,
scattered ephemeral puddles or standing water in flooded fields, whereas
cricket frogs call from larger and more permanent bodies of water such as
ponds and streams. Examination of a much greater number of species, performed
in a way that is sensitive to phylogenetic relationships, is necessary to
extract the rules by which species become more specialized for long-distance
or short-distance communication.
Although we cannot determine the evolutionary forces dictating the specializations of any one species, our results do address a neglected issue in the behavioral ecology and evolution of communication system—the tension between the requirements of long and short distance communication. We propose that an acoustic communication system using the same signal for communication over a variety of distances can be specialized for either long-distance or short-distance communication. It cannot, however, be equally sensitive to distant and to close calls in any environment in which there is noticeable signal degradation during transmission. Each species must therefore effect a compromise between long- and short-distance communication, and favoring one results in a disadvantage in the other.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
We thank N. Suga for discussions that led to this study, K. Beeman for technical advice on Signal, and C. Gerhardt, N. Kime, J. Schwartz, and K. Witte for comments on the manuscript. This work was supported by National Science Foundation grants IBN-93-16185 (to M.J.R. and A.S.R.), BNS 86-02689, 90-021185 (to M.J.R. and W.W.), National Institutes of Health (NIH) grant RO1 MH52696 (to W.W. and M.J.R.), the Smithsonian Scholarly Studies Program (A.S.R. and M.J.R.), and Central Washington University (L.S.). All experiments reported here comply to the NIH's Principle of Animal Care (no. 86-23, revised 1985).
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Andersson M, 1994. Sexual selection. Princeton, New Jersey: Princeton University Press.
Beeman K, 1994. Signal. Belmont, Massachusetts: Engineering Design.
Blair WF, 1958. Mating call in the speciation of anuran amphibians. Am Nat 92:27 -51.[Web of Science]
Bourne GR, 1993. Proximate costs and benefits of mate acquisition at leks of the frog Ololygon rubra. Anim Behav 45:1051 -1059.
Brenowitz EA, Wilczynski W, Zakon HH, 1984. Acoustic communication in spring peepers: Environmental and behavioral aspects.J Comp Physiol 155:585-593.
Capranica RR, 1976. Auditory system. In: Frog neurobiology (Llinas R, Precht W, eds). Berlin: Springer-Verlag;551 -575.
Capranica RR, Frischkopf LS, Nevo E, 1973. Encoding of
geographic dialects in the auditory system of the cricket frog.Science
182:1272
-1275.
Christy JH, 1995. Mimicry, mate choice, and the sensory trap hypothesis. Am Nat 146:171-181.[Web of Science]
Darwin C, 1871. The descent of man and selection in relation to sex. New York: Random House.
Dawkins MS, Guilford T, 1996. Sensory bias and the adaptiveness of female choice. Am Nat 148: 937-942.
Dusenbery DB, 1992. Sensory ecology. New York: Freeman.
Eberhard WG. 1993. Evaluating models of sexual selection: genitalia as a test case. Am Nat 142:564-571.
Endler JA, 1992. Signals, signal conditions, and the direction of evolution. Am Nat 131:S125-S153.
Feng AS, Hall JC, Gooler DM, 1990. Neural basis of sound pattern recognition in anurans. Prog Neurobiol 34:313-329.[Web of Science][Medline]
Forrest TG, 1994. From sender to receiver: propagation and environmental effects on acoustic signals. Am Zool 34:644-654.
Fox JH, 1988. Possible active space optimization by the male cricket frog, Acris crepitans, through adjustment of neighbor spacing. Soc Neurosci Abstr 14:88.
Fox JH, 1995. Morphological correlates of auditory sensitivity in anuran amphibians. Brain Behav Evol 45:327-338.[Web of Science][Medline]
Fritzsch B, Ryan M, Wilczynski W, Hetherington T, Walkowiak W (eds), 1988. The evolution of the amphibian auditory system. New York: John Wiley and Sons.
Gerhardt HC, 1994a. The evolution of vocalization in frogs and toads. Annu Rev Ecol Syst 25:293-324.[Web of Science]
Gerhardt HC, 1994b. Selective responsiveness to long-range acoustic signals in insects and anurans. Am Zool 34:706-714.
Guilford T, Dawkins MS, 1993. Receiver psychology and the design of animal signals. Trends Neurosci 16:430-436.[Web of Science][Medline]
Hall JC, 1994. Central processing of communication sounds in the anuran auditory system. Am Zool 34:670-680.
Howard RD, Whiteman HH, Schueller TI, 1994. Sexual selection in American toads: a test of a good-genes hypothesis.Evolution 48:1286 -1300.[Web of Science]
Keddy-Hector, Wilczynski W, Ryan MJ, 1992. Call patterns and basilar papilla tuning in cricket frogs. II. Intrapopulational variation and allometry. Brain Behav Evol 39:238-246.[Web of Science][Medline]
Kirkpatrick M, Ryan MJ, 1991. The paradox of the lek and the evolution of mating preferences. Nature 350:33-38.
Littlejohn MJ, Michaud T, 1959. Mating call discrimination by females of Strecker's chorus frog (Pseudacris streckeri). Texas J Sci 11:86-92.
Mayr E, 1963. Animal species and evolution. Cambridge, Massachusetts: Harvard University Press.
Maynard Smith J, Burian R, Kauffman S, Alberch P, Campbell J, Goodwin B, Lande R, Raap D, Wolpert L, 1985. Developmental constraints and evolution. Q Rev Biol 60:265-287.
McClelland BE, Wilczynski W, Rand AS, 1997. Sexual dimorphism and species differences in the neurophysiology and morphology of the acoustic communication system of two neotropical hylids. J Comp Physiol 180:451-462.
Nevo E, Capranica RR, 1985. Evolutionary origin of ethological reproductive isolation in cricket frogs, Acris.Evol Biol 19:147-214.
Rand AS, 1988. An overview of anuran acoustic communication. In: The evolution of the amphibian auditory system (Fritzsch B, Ryan M, Wilczynski W, Hetherington T, Walkowiak W, eds). New York: John Wiley and Sons;415 -431.
Rand AS, Ryan MJ, 1981. The adaptive significance of a complex vocal repertoire in a neotropical frog. Z Tierpsychol 57:209-214.
Rand AS, Ryan MJ, Wilczynski W, 1992. Signal redundancy and receiver permissiveness in acoustic mate recognition by the túngara frog, Physalaemus pustulosus.Am Zool 32:81-90.
Robertson JG, 1990. Female choice increases fertilization success in the Australian frog, Uperoleia laevigata.Anim Behav 39:639-645.
Ryan MJ, 1980. Female mate choice in a neotropical
frog. Science
209:523-525.
Ryan MJ, 1983. Sexual selection and communication in a neotropical frog, Physalaemus pustulosus. Evolution 37:261-272.
Ryan MJ, 1985. The túngara frog: a study in sexual selection and communication. Chicago: University of Chicago Press.
Ryan MJ, 1986. Environmental bioacoustics: evaluation of a commonly-used experimental design. Anim Behav 34:931-933.
Ryan MJ, 1990. Sensory systems, sexual selection, and sensory exploitation. Oxford Surv Evol Biol 7:157-195.
Ryan MJ, 1991. Sexual selection and communication in frogs. Trends Ecol Evol 6:351-355.
Ryan MJ, 1994. Mechanistic studies in sexual selection. In: Behavioral mechanisms in evolutionary ecology (Real L, ed). Chicago: University of Chicago Press;190 -215.
Ryan MJ, 1997. Sexual selection and mate choice. In:Behavioural ecology: an evolutionary approach (Krebs JB, Davies NB, eds). Cambridge: Blackwell; 179-202.
Ryan MJ, Cocroft RB, Wilczynski W, 1990a. The role of environmental selection in intraspecific divergence of mate recognition signals in the cricket frog, Acris crepitans.Evolution 44:1869-1872.
Ryan MJ, Fox JH, Wilczynski W, Rand AS, 1990b. Sexual selection for sensory exploitation in the frog Physalaemus pustulosus. Nature 343:66-67.[Medline]
Ryan MJ, Keddy-Hector A, 1992. Directional patterns of female mate choice and the role of sensory biases. Am Nat 139:S4-S35.
Ryan MJ, Perril SA, Wilczynski W, 1992. Auditory tuning and call frequency predict population-based mating preferences in the cricket frog, Acris crepitans. Am Nat 139:1370-1383.
Ryan MJ, Rand AS, 1995. Female responses to ancestral
advertisement calls in tungara frogs. Science
269:390-392.
Ryan MJ, Wilczynski W, 1988. Coevolution of sender and
receiver: effect on local mate preference in cricket frogs.Science
240:1786-1788.
Ryan MJ, Wilczynski W, 1991. Evolution of intraspecific variation in the advertisement call of a cricket frog (Acris crepitans, Hylidae). Biol J Linn Soc 44:249-271.
Schwartz JJ, Gerhardt HC, 1998. The neuroethology of frequency preferences in the spring peeper. Anim Behav 56:55-69.[Web of Science][Medline]
Shaw K, 1995. Phylogenetic tests of the sensory exploitation model of sexual selection. Trends Ecol Evol 10:117-120.
Shine R, 1979. Sexual selection and sexual dimorphism in the amphibia. Copeia 1979:297-306.
West-Eberhard MJ, 1979. Sexual selection, social competition, and evolution. Proc Am Phil Soc 123:222-234.[Web of Science]
Wilczynski W, 1986. Sexual differences in neural tuning and their effect on active space. Brain Behav Evol 28:83-94.[Web of Science][Medline]
Wilczynski W, Capranica RR, 1984. The auditory system of anuran amphibians. Prog Neurobiol 22:1-30.[Web of Science][Medline]
Wilczynski W, Keddy-Hector A, Ryan MJ, 1992. Call patterns and basilar papilla tuning in cricket frogs. I. Differences among populations and between sexes. Brain Behav Evol 39:229-237.[Web of Science][Medline]
Wilczynski W, McClelland BE, Rand AS, 1993. Acoustic auditory, and morphological divergence in three species of neotropical frog.J Comp Physiol 172:425-438.[Medline]
Wilczynski W, Rand AS, Ryan MJ, 1995. The processing of spectral cues by the call analysis system of the túngara frog, Physalaemus pustulosus.Anim Behav 49:911-929.
Wilczynski W, Ryan MJ, 1999. Geographic variation in animal communication systems. In: Geographic variation in behavior: an evolutionary perspective (Foster SA, Endler J, eds). Oxford: Oxford University Press; 234-261.
Wilczynski W, Zakon HH, Brenowitz EA, 1984. Acoustic communication in spring peepers, call characteristics and neurophysiological aspects. J Comp Physiol 155:577-584.
Wiley RH, Richards DG, 1978. Physical constraints on acoustic communication in the atmosphere: implications for the evolution of animal vocalizations. Behav Ecol Sociobiol 3:69-94.
Zakon HH, Wilczynski W, 1988. The physiology of the VIII nerve. In: The evolution of the amphibian auditory system (Fritzsch B, Ryan MJ, Wilczynski W, Hetherington T, Walkowiak W, eds). New York: John Wiley and Sons; 125-155.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  S. M Phelps Sensory ecology and perceptual allocation: new prospects for neural networks Phil Trans R Soc B, March 29, 2007; 362(1479): 355 - 367. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Witte, H. E. Farris, M. J. Ryan, and W. Wilczynski How cricket frog females deal with a noisy world: habitat-related differences in auditory tuning Behav. Ecol., May 1, 2005; 16(3): 571 - 579. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. Heiling, K. Cheng, and M. E. Herberstein Exploitation of floral signals by crab spiders (Thomisus spectabilis, Thomisidae) Behav. Ecol., March 1, 2004; 15(2): 321 - 326. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||