Behavioral Ecology Vol. 13 No. 5: 615-621
© 2002 International Society for Behavioral Ecology
Sexual selection in the wolf spider Hygrolycosa rubrofasciata: female preference for drum duration and pulse rate
Department of Biological and Environmental Science, University of Jyväskylä, PO Box 35, Jyväskylä, Finland
Address correspondence to S. Parri. E-mail: pasika{at}cc.jyu.fi. A. Rivero is now at the Institute of Cell, Animal and Population Biology, University of Edinburgh, Kings Buildings, West Mains Road, Edinburgh EH9 3JT, UK.
Received 16 January 2001; revised 12 December 2001; accepted 18 December 2001.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The unusual form of sexual signaling, the drumming produced by the wolf spider Hygrolycosa rubrofasciata, allows exceptionally detailed studies of female preference patterns against signal characteristics. It is easy to manipulate the signals and to use large numbers of females in playback experiments. Males of H. rubrofasciata produce drums by striking their abdomen against dry leaves on the ground. Drums travel not only as substrate-borne vibrations, but also as airborne acoustic signals. Females respond sooner to drums transferred as substrate borne, but the mode of signal transfer has no effect on female preference for different types of drums. We investigated the effects of two key components of the drums, duration and pulse rate, on female choice. Previous studies have shown that the duration of the drum is both repeatable within males and variable between males. In contrast, pulse rate shows high within-male repeatability but only little variability between males. Using playbacks of manipulated drums, we found that females preferred drums of longer duration but that pulse rate had no effect on female preference. Our results suggest that drum duration may function as an indicator of male quality for choosy females. Pulse rate, on the other hand, is less likely to be an important trait in intersexual selection. Female preference for drum duration was open-ended within the natural variation of the drum durations, but it leveled off outside the normal range. Thus supernomal stimuli would not pay for males using this energetically demanding acoustic signal.
Key words: acoustic signaling, drum duration, female choice, Hygrolycosa rubrofasciata, pulse rate, sexual selection, wolf spiders.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Acoustic signals are particularly suitable for experiments of female preference against signal properties that can be manipulated for playbacks. Female mate choice based on male acoustic signals is driven by various components or characteristics of the signals, such as calling rate, duration, intensity, dominant frequency, and song complexity (see review in Ryan and Keddy-Hector, 1992
;
see also Brown et al., 1996;
Catchpole, 1987,
1996;
Cherry, 1993;
Eiríksson, 1993;
Gray, 1997;
Ritchie et al., 1998;
Searcy and Andersson, 1986;
Sullivan and Hinshaw, 1992).
In invertebrates the signal components most often involved in female choice
are frequency, duration, and pulse rate. Signal duration and pulse rate are
important for female choice, for example, in orthopteran calls
(Butlin et al., 1985;
Eiríksson, 1993;
Hedrick, 1986;
Simmons, 1988) but also in
some other insects (Jang and Greenfield,
1996), and even in spiders
(Schmitt et al., 1993).
Typically, females prefer male signals of greater magnitude (e.g., longer
duration, greater repetition rates and larger repertoires;
Baker et al., 1986; for
reviews, see Andersson, 1994;
Ryan, 1988). This may be the
result of females choosing signals of greater energy content, which in turn
may imply that females use these signals as indicators of male quality.
Indeed, there is evidence that sexual traits function as indicators of
indirect genetic benefits for females in terms of good genes (e.g.,
Alatalo et al., 1998;
Hasselquist et al., 1996;
Hoikkala et al., 1998;
Welch et al., 1998; for
review, see Møller and Alatalo,
1999). For example, in Drosophila montana, females
benefit in terms of better survival of their offspring by choosing males that
produce short signals with a high carrier frequency
(Hoikkala et al., 1998).
Females may also gain direct benefits from their preference for male sexual
traits. For example, in the black horned tree cricket, Oecanthus
nigricornis, male song correlates with body size and age, and females
mating with larger males benefit directly through increased fecundity
(Brown et al., 1996).
Studies of spider communication have mainly investigated the role and
importance of the substrate-borne vibrations (e.g.,
Baurect and Barth, 1992;
Rovner and Barth, 1981;
Schmitt et al., 1993,
1994;
Schüch and Barth, 1990).
Only a few studies in spiders have even considered airborne acoustic
communication (Barth, 1982;
Harrison, 1969;
Rovner, 1967;
Stratton and Uetz, 1981;
Uetz and Stratton, 1982).
Further-more, the acoustic signaling of spiders has rarely been studied from
the viewpoint of indicator mechanism of sexual selection (studied by us on a
single species as an exception; see Alatalo
et al., 1998; Kotiaho,
2000; Kotiaho et al.,
1996,
1998,
1999;
Mappes et al., 1996;
Rivero et al., 2000).
In the wolf spider, Hygrolycosa rubrofasciata (Ohlert), male
sexual display consists of a series of drums produced by striking the abdomen
against dry leaves on the ground. The natural habitat of the spider is open
bogs and fields, and the signals are audible to humans up to a distance of 10
m. At the time of the mating season, the habitat is covered with layers of
dry, deciduous leaves and dry grass. When searching for receptive females,
males patrol around the habitat and stop to drum on the dry uppermost leaves
on the ground (Kotiaho et al.,
2000). In this complex litter habitat, drums travel as vibrations
in the substrate but also as audible, airborne vibrations. Airborne and
substrate-borne vibrations have basically the same structure and thus in
principle convey the same information. In spiders the same sensory organs,
slit sensilla, are used to detect both types of vibrations
(Foelix, 1996;
Hergenröder and Barth,
1983). In addition, spiders have sensilla (termed trichobothria)
specialized in detecting air movements and low-frequency airborne vibrations
(Foelix, 1996). The frequency
of the drums range between 1 and 8 kHz, thus being clearly audible for the
human ear (Rivero et al.,
2000).
A detailed study of the characteristics of the drum of H.
rubrofasciata males has been made
(Rivero et al., 2000). This
study identified two main components of the male drum, based on the within-
and between-male variability: drum duration and pulse rate. Drum duration is
moderately repeatable within males and considerably variable between males,
while pulse rate (number of pulses divided by drum duration) is more static,
having high repeatability within males but very small variation between the
males (Rivero et al., 2000).
In addition, drum duration is positively correlated with drumming rate (as
number of drums per minute), whereas pulse rate is not
(Rivero et al., 2000).
Previous studies have shown that drumming rate is highly repeatable and
indicates both male phenotypic (Kotiaho,
2000; Kotiaho et al.,
1996; Mappes et al.,
1996) and genetic quality
(Alatalo et al., 1998).
Furthermore, there is evidence that females prefer to mate with males of high
drumming rate (Kotiaho et al.,
1996; Parri et al.,
1997).
In this study we first examined the unusually audible spider signal in
respect to the mode of the transfer of male signals (substrate-borne versus
airborne alone) and female preference. Second and most important, we examined
the effect of drum duration and pulse rate on female mate choice using
manipulated signals in playbacks. We expected that drum duration, as the more
dynamic trait, would be more important for female preference. In particular,
we examined the shape of female preference against the size of the male
signal. The-oretical models have often assumed gaussian shapes for preferences
against male trait (e.g., in the classic Fisherian models;
Lande, 1981). However, the
shapes of the preference curves have seldom been examined in any previous
experiments (but see, e.g., Eiríksson,
1993,
1994;
Gerhardt, 1991;
Gray and Cade, 1999;
Jang and Greenfield, 1996)
because typically there has been only one category with increased or reduced
ornaments.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The species and the signal
The study species Hygrolycosa rubrofasciata is a ground-dwelling wolf spider (Lycosidae). Typical habitat of this species is open bogs and meadows where the spider occurs in patchily distributed populations. In southern Finland the mating season begins soon after snowmelt, typically toward the end of April. During this period, males move around the habitat, occasionally stopping to drum. When a male detects a female, he stops and increases his drumming rate. If the female is willing to mate with the male, she responds to the male drum by vibrating her body. This response takes place immediately after the male drum, thus making it easy to determine which male the female responded to. Before the mating the male and the female both produce a few drums while the male approaches the female ("duetting"; Kronestedt, 1996
). Female response is essential for the mating to begin.
Contrary to other lycosids (for review, see
Forster, 1982), the vision of
H. rubrofasciata is poor, and it is not used in mating (Parri et al.,
unpublished observations).
Males produce drums by striking their abdomen against dry, deciduous leaves
(Kotiaho et al., 2000;
Kronestedt, 1984,
1996). Males will also drum on
other materials, such as paper
(Kronestedt, 1984). In the
laboratory, males drum on average 0.8 times/min (SD ± 0.7, n =
423; Rivero et al., 2000).
Each drum lasts on average 1 s and consists of a series of pulses. Each pulse
corresponds to a single strike of the abdomen
(Figure 1). The mean pulse rate
of the drum is 0.029 pulses/ms (Rivero et
al., 2000). The relatively simple structure of the drum gives us
an opportunity to manipulate specific characteristics of the drum and to study
female preference for these characteristics in detail.
|
Collecting the spiders from the field
All experiments were conducted using virgin individuals of H.
rubrofasciata. We collected spiders in late April 1995 and 1996 from two
populations, Stormossa and Stenberg, in southern Finland. Spiders were placed
individually into small, plastic containers (diam 3.4 cm, height 6.7 cm) with
moist moss (Sphagnum sp.) to maintain humidity and brought into the
laboratory to be weighed. Spiders were maintained in darkness at
4°-8°C to prevent them from reaching full sexual activity. Two or
three days before the experiments spiders were taken to the room temperature,
20°-24°C. Spiders were fed fruit flies, Drosophila sp.
Mode of signal transfer
In the signal transfer experiment we tested whether the mode of male signal
transfer (substrate borne versus airborne alone) affects female preference.
The experiment was conducted in two setups on a 30 x 30 cm arena, the
bottom of which was covered with 2.5-cm thick foam. In the first setup
(contact setup), a paper cross with four arms of equal length was placed on
top of the foam. On each of the arms we marked positions for male cages at
distances of 15, 10, and 5 cm from the center. On the center of the cross we
marked a position for a female cage. The second setup (no-contact setup) was
identical to the first one, except that the paper and the foam under the
female cage were cut loose from the rest of the arena. Male and female cages
were bottomless, 5.0-cm tall cylinders made out of sheets of clear acetate
with a radius of 3.4 cm. In the contact setup, the intact paper cross allowed
a direct substrate-borne (and airborne) contact between the female and the
males, while in the no-contact setup direct substrate-borne contact did not
exist. These two setups were motivated by our observations that typically in
the field the leaves on which males drum are not in direct contact with the
leaves on which females stand. Although there are two modes of signal
transfer, airborne and substrate borne, it is still possible that females
receive the signals with the help of the substrate in spite of the airborne
transfer. The airborne signals may transfer into the substrate, allowing the
ultimate sensory stimulus to be received as substrate borne rather than
airborne. Our aim was not to study the mechanisms of receiving the airborne
signals by females, but rather to determine whether the mode of signal
transfer affects female preference.
Before the experiment we placed male cages at the outer-most positions on each of the four arms (i.e., 15 cm away from the center). To stimulate male drumming, a female was placed into each of the male cages for 10 min and removed before the beginning of the trial. After removing the females, males were placed into the male cages. Once each of the males had started to drum, the trial was begun by placing an experimental female into the female cage at the center of the arena. We observed female behavior for 10 min. If the female did not respond during the first 10 min, male cages were carefully moved closer to the female to the distance of 10 cm. If the female did not respond during the next 10 min, we moved the males once more, to 5 cm from the female. If there was no response during the third 10 min, the trial was terminated and discarded. We avoided placing the male at the closest distance from the start because preliminary trials had shown us that at short distances females can become overresponsive. Once a female responded, we noted the time of the response, noted to which male the female responded, and ended the trial. The duration of the trials thus varied from a few seconds to 30 min. In occasional cases where there were two males drumming so close in time to each other that it was impossible to determine whom the female responded to, the trial was discarded.
We used males and females from two populations separated by 2 km of forest.
Trials were replicated 40 times with Stormossa females (20 contact and 20
no-contact setups) and 40 times with Stenberg females (20 contact and 20
no-contact setups). In each of the trials two of the males came from Stormossa
and two from Stenberg. We used each male and each female only once. We
compared female responsiveness between the two populations and found no
difference in either time until female response (Stormossa: mean ± SD =
5.24 ± 7.45 min; Stenberg: 3.70 ± 4.60 min; Mann-Whitney
Z = 0.24, n = 29, 27, respectively, p = .807) or in
the proportion of trials where the female responded (Stormossa: 29/40;
Stenberg: 27/40; Yates's corrected 
2 = 0.06, df = 1,
n = 80, p = .807). Most important, females did not
discriminate between the males from the two populations (Yates's corrected
2 = 0.50, df = 1, n = 51, p = .478). For
this reason, all subsequent analyses were carried out with data that are
pooled over the two populations.
Drum duration
For the drum duration experiment we recorded male drums using a digital
recorder (Casio DA-7) attached to a Telinga microphone. For the recordings,
males were placed individually in a plastic arena (diam 6.5 cm, height 4 cm)
with some dry birch leaves as drumming substrate. For further details on
recordings, see Rivero et al.
(2000). The recordings were
manipulated with sound-editing programs in a computer.
We used a total of 37 drums, each from a different male. Each drum was manipulated into four different durations by decreasing the original duration by 25% (D—25) and 50% (D—50) and by increasing the original duration by 25% (D+25) and 50% (D+50). To maintain the original structure of the drums (Figure 1), we decreased the drum duration by removing every second (D—50) or every fourth pulse (D—25), taking care that the temporal structure of the drum (i.e., the time between successive pulses) did not change. Similarly, the duration of the drums was increased by duplicating every second (D+50) or every fourth (D+25) pulse. We used the original drum as a control for the manipulated drums. To account for the possible artifacts resulting from removing and adding pulses within the drum, the control drums were also treated by removing pulses and adding them back into their original positions. We used each original drum with its four manipulated variants to create a playback loop on the computer. Thus, each of the 37 loops consisted of 5 drums arranged in random order. The loops were replayed with a repetition rate of 4 drums/min, and each of the loops constituted one replicate.
The loops were replayed with a computer and four loudspeakers (Pioneer TS-F
1665). As a replicate we used a group of females (for groups, see below) that
was divided to four loudspeakers and replayed a loop simultaneously. Females
were in individual acetate cages described earlier but now with paper bottom,
and thus it was possible for two observers (two loudspeakers per observer) to
follow the females individually. It was possible to fit five female cages on
one loudspeaker. The loops were replayed to groups of females instead of
individually to increase the effectiveness of the experimental procedure; it
would have been very laborious to replay a loop to each female individually.
For results we calculated the proportion of responses given for each drum
duration within each female group. These proportions were used in the final
analysis. From previous experiments, we know that females are sensitive to the
volume of the drums (Parri et al.,
1997). Therefore, to elicit responses from as many females as
possible and to allow females to compare between the drums, we began each
playback session with a low volume. Each loop was replayed three rounds to the
group of females at low volume. The volume was then increased and the loop
replayed three more rounds. The volume was increased once more so that each
group of the females was replayed a total of 45 drums, 15 on each of the 3
volume levels. While replaying the loops to the females we noted to which of
the drums each female responded for the first time.
We used the general experimental design described above in two experiments: one with contact-only setup and the other with both contact and no-contact setup. In the contact-only setup, females were placed directly on the loudspeakers. We replayed 11 loops to a group of 16-20 females (4 or 5 females per loudspeaker) using each female only once. These females originated from the Stormossa population. The second setup, contact and no-contact setup, was conducted to determine whether the signal transfer mode (substrate and airborne versus airborne alone) affected female preference for the drum duration. We placed 26 groups of 10 females either directly on the loudspeaker (contact setup) or on a metal grid 3 cm above the loudspeaker (no-contact setup). Each female group was replayed a loop of manipulated drums once on contact and once on no-contact setup. The test was conducted in 2 successive days, and the order of contact and no-contact setups was random. These females originated from the Stenberg population. The size of female groups between the setups (16-20 and 10 females) varied because of the difference in total number of females that were available for the experiments.
We tested whether the females from the two populations differed in their preference for drum duration by comparing the proportion of females, within each replicate, that was responding to each of the drum durations. This comparison was possible to conduct in the contact setups of the above two experiments. To increase the statistical power of the tests, we pooled the proportions of females responding to decreased drum duration (D—25 and D—50) and to increased drum duration (D+25 and D+50). There were no differences between the two populations at any of the three drum durations (for decreased drum duration Mann-Whitney U = 123, for control drums U = 109 and for increased drum duration U = 126, for each comparison n = 11, 26, respectively; p > .255). Thus, for further analysis of the shape of female preference in the contact setup, we pooled the populations.
Pulse rate
We manipulated 15 drums from Stormossa males that were recorded in 1995.
The mean (± SD) original pulse rate of these drums was 0.028 ±
0.0012 pulses/ms. The mean (± SD) pulse rate within the natural
population in 1995 was 0.029 ± 0.0017 pulses/ms (n = 78)
(Rivero et al., 2000). The
pulse rate of the drums was either decreased or increased by 10% (P—10
and P+10, respectively) by manipulating the inter-pulse interval. To retain
the original duration of the drums, we either added or removed individual
pulses from the drums. We used the original drum as control and manipulated it
as above without changing the interpulse interval. Each control drum and its
two manipulated forms were used to create a playback loop. In each of the 15
loops, the order of the drums was random, and the drums were repeated at a
rate of 4 drums/min. Each loop was replayed simultaneously to a group of 16-20
Stormossa females divided to four loudspeakers. Each group was used as a
replicate, and each female was used only once. The playback procedure was
identical to the one used in the contact setup of the drum duration
experiment. We increased playback volume twice after replaying each loop five
times (i.e., always after 15 drums). As in the drum duration experiment, we
recorded which drum each female responded to for the first time and calculated
the proportion of responses for each manipulation level within each female
group. We tested the results of all experiments using nonparametric tests
because the assumptions of normality of the data were not met.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Mode of signal transfer
The proportion of the trials where the female responded to drums did not differ between the contact and no-contact setups (28/39 and 28/41, respectively; Yates's corrected
2 = 0.01, n = 80,
p = .922). However, the time until the female responded was
significantly shorter in the contact than in the no-contact setup (mean
± SD = 2.18 ± 2.13 and 6.79 ± 7.80 min, respectively;
Mann-Whitney z = 2.82, n = 28 for both setups, p =
.005). In the contact setup, all females responded to a male already from the
longest distance (15 cm). In the no-contact setup, 23 females out of the 28
responded from 15 cm, while for the other females, males needed to be moved
within 10 or even 5 cm from the female
(Figure 2).
|
Drum duration
First, to test female preference for drum durations as effectively as
possible, we collapsed the five levels of drum durations into three levels by
pooling the female responses to the two decreased drum durations and by
pooling the female responses to the two increased drum durations. This was
done by using means of each combined group. The proportion of the responses
given in the contact setup differed significantly between the three durations
(Friedman 2 = 28.93, df = 2, n = 26, p <
.001, Figure 3). Females
preferred the original duration over the decreased ones (Wilcoxon z =
3.56, n = 26, p < .001, probability level is adjusted for
multiple comparisons with sequential Bonferroni correction
[Rice, 1989]), and there was
also a tendency for preference for increased duration over the original
duration (Wilcoxon z = 1.89, n = 26, p = .059,
probability level is adjusted for multiple comparisons with sequential
Bonferroni correction). The difference between decreased and increased drum
durations was significant (Wilcoxon z = 4.34, n = 26,
p < .001, probability level is adjusted for multiple comparisons
with sequential Bonferroni correction;
Figure 3).
|
There was also a significant difference between female preference for drum
durations in the no-contact setup (Friedman 2 = 7.86, df = 2,
p = .020, Figure 3). A
posteriori test revealed that the proportion of responses differed
significantly between decreased and increased durations (Wilcoxon z =
2.77, n = 26, p < .018, probability level is adjusted for
multiple comparisons with sequential Bonferroni correction;
Figure 3), but not between
decreased and original or original and increased durations (Wilcoxon
z = 1.69, n = 26, p = .184, and z = 0.61,
n = 26, p = .543, respectively, probability levels are
adjusted for multiple comparisons with sequential Bonferroni correction).
To test for female preference between the contact and no-contact setups, we calculated an average duration to which each of the female groups responded. This was done by assigning a value of 0.5 for each response to decreased drum duration, 1 for each response to control drum and 1.5 for each response to increased drum duration. The resulting mean durations were tested with a rank order test. There was no significant difference in female preference for drum duration between the contact and no-contact setups (Wilcoxon z = 1.59, n = 26, p = .113). Nevertheless, in the contact setup drums elicited female responses sooner than in the no-contact setup (mean number of drums before response ± SD = 14.90 ± 6.10, and 25.41 ± 6.69, respectively; n = 26, Wilcoxon z = -3.92, p < .001 [as the drums were replayed in regular intervals of 15 s the number of the drums before response is equivalent to the time until response]). Thus, the mode of the signal transfer had no effect on the female preference for drum duration, but it affected the timing of the response.
Finally, to study the selection that female preference imposes on drum
duration, we applied a curve-fitting analysis to the data from the contact
setup of the drum duration experiments. The best fit to the data was given by
a cubic regression (r2 = .29, F3,181 =
24.84, p < .001; b1 = -3.59 ± 1.59, t =
-2.25, p < .026; b2 = 4.22 ± 1.70, t =
2.49, p < .014; b3 = -1.41 ± 0.56, t =
-2.51, p < .013). The cubic function is depicted in
Figure 4 over the bars of the
proportion of responses on each drum duration. Below the bars is the
distribution of drum durations drawn from the field population in the same
year (mean ± SD = 981.4 ± 140.6 ms, range = 681.0-1329.7,
n = 78; Rivero et al.,
2000). The mean duration of the drums that were used for the
manipulation was 1077.5 ± 184.8 (SD) ms (n = 37). By comparing
the female preference function to the distribution of drum durations drawn
from the field population, it can be concluded that female preference is for
greater than the natural mean drum duration. However, the preference is not
completely open-ended, but instead the preference for even longer drums levels
off after about 25% increase in drum duration.
|
Pulse rate
The pulse rate manipulation had no effect on female preference (Friedman
2 = 0.79, df = 2, p = .673;
Figure 5). To further compare
female preference for the pulse rate as powerfully as possible, we scored
every response given to the decreased pulse rate with a negative value (-1),
and every response given to the increased pulse rate with a positive value
(+1). Using the scores calculated for each loop separately, we calculated the
mean direction for female preference within each trial. Mean directions were
then compared against zero (i.e., the situation where there is no directional
selection by the females). Comparison indicated that there was no directional
female preference for pulse rate (mean of all the trials ± SD = 0.034
± 0.048, Wilcoxon T = 39, n = 14, p =
.320).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
In the first experiment with live males, we found that signal transfer mode had an effect on female responsiveness such that females responded sooner in the contact setup than in the no-contact setup. The playback experiment on drum durations rendered the same result: females in the contact setup responded sooner than females in the no-contact setup. However, despite the delayed response in no-contact setup, the mode of signal transfer did not alter female preference for the duration of drums. In other words, when the signal is transferred only via the air, it takes longer for the female to choose between the signals, but the final preference for the signal type does not change. It is likely that the contact setups elicited faster responses from the females because the information content of the drum is more effectively transferred when in direct contact to the source of the signal. This result parallels the one found earlier with manipulations of signal volume; females respond sooner to signals replayed at higher volumes (Parri et al., 1997
).
The experiment on signal duration revealed that females of H.
rubrofasciata prefer increased drum duration over the decreased drum
duration. In general, signal duration is often found to be under directional
sexual selection by the females (but see also
Butlin et al., 1985;
Eiríksson, 1993). For
example, females of the field cricket Gryllus integer use calling
bout duration as a criterion in their mate choice, and females predominantly
choose calls of greater duration (Hedrick,
1986). Another example is the lesser wax moth Achroia
grisella, in which females choose pulse durations over the population
mean (Jang and Greenfield,
1996).
Honest signaling theories of sexual selection (see, e.g.,
Andersson, 1994;
Grafen, 1990;
Johnstone, 1995) suggest that
females could use the magnitude of male advertisement as an indicator of male
phenotypic or genetic quality. According to this theory, females may be
expected to show preference for greater magnitudes of male signals because
males of poor quality cannot afford to have or maintain signals of relatively
great magnitude. Such female preference may lead to an open-ended directional
selection for the greater magnitude of male sexual traits.
We found that females preferred the increased drum durations, which were
clearly greater than the natural population mean. In fact, female preference
increased steeply with drum duration across the natural range of drum duration
variation. An asymptotic level of female preference was approached after a 25%
increase in drum duration. Only few natural drums reach the duration of this
magnitude. Thus, it seems that females do not exhibit an open-ended preference
for supernormal stimuli, but the preference levels off with the natural
signals of greatest duration. Because drumming is energetically very expensive
to males (Kotiaho et al.,
1998), the benefits for males of using very long drums must
decline steeply with drum duration.
Earlier, we found that drum duration shows high variability between males
(CVs 14-30%) and high repeatability within males (0.55-0.66;
Rivero et al., 2000). In
addition, drum duration is correlated with another key component of the male's
sexual advertisement, its drumming rate
(Rivero et al., 2000). In
other studies we found that females of H. rubrofasciata prefer high
drumming rate (Kotiaho et al.,
1996; Parri et al.,
1997), which is an honest indicator of male phenotypic but also of
heritable genetic quality (Alatalo et al.,
1998; Kotiaho,
2000; Mappes et al.,
1996). Thus, it is likely that females gain information about male
quality via drum duration, explaining the strong directional female preference
for higher drum durations.
Pulse rate manipulation did not affect female preference. In a previous
study, we found that between-male variability for pulse rate was low (CV
5-7%), while within-male repeatability was relatively high (0.62-0.84;
Rivero et al., 2000). Because
quality indicator traits that are under directional female preference
generally show both high variability within but also between males
(Pomiankowski and Møller,
1995), it is plausible that pulse rate with low variability
between males is not an important indicator trait. In several other studies,
pulse rate of male signals has been found to be under stabilizing rather than
directional selection. For example, in a playback experiment with synthetic
advertisements calls females of the natterjack toad Bufo calamita
showed preference for the intermediate pulse rate
(Arak, 1988; see also
Gray and Cade, 1999).
Typically, indicator traits that honestly reflect male quality incur costs
to the bearer of the trait. Several studies indicate that acoustic calls are
among the most energetically demanding male sexual traits (e.g.,
Kotiaho, 2001;
Prestwich, 1994;
Vehrencamp et al., 1989).
Indeed, there is some evidence that the energy requirements of signaling and
male attractiveness to females might be correlated with each other: In the wax
moth Achroia grisella, energy expenditure in signaling was highest in
most attractive males (Reinhold et al.,
1998). In H. rubrofasciata, energy consumption of
drumming is, on average, 22 times higher than the energy consumption during
resting (Kotiaho et al.,
1998). Although the relative energy contents of drums of varying
durations have not been investigated, it is likely that drums of longer
duration involving a larger number of pulses are energetically more demanding
to produce and thus more costly.
In conclusion, signal transfer mode had an effect on female responsiveness such that females responded sooner in the contact setup than in the no-contact setup. However, mode of signal transfer did not alter female preference for the duration of drums. Pulse rate of the drums did not affect female preference. Females did, however, show clear preference for signal duration greater than its natural mean. Because females prefer to mate with high-drumming-rate males who are in better phenotypic condition and drumming rate is correlated with signal duration, it may be that females of H. rubrofasciata are able to use male signal duration as an indicator of male quality. It is interesting that female preference was not entirely open ended but leveled off after signal durations above natural variation.
|
ACKNOWLEDGEMENTS |
|---|
This work was funded by the Academy of Finland as a research grant and by the European Union as an HCM Network grant to R.V.A. We thank Konnevesi Research Station and the Experimental Animal Unit for providing us good working facilities and for help in maintaining the spiders. Jaana Suutari, Tapio Tikkanen, Eevi Karvonen, Kaisa Rikalainen, Saija Hytönen, and Perttu Rantanen helped us collect the spiders in the field and conduct the experiments in the laboratory. Marlene Zuk and Veijo Jormalainen provided us with helpful comments on earlier versions of the manuscript.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Alatalo RV, Kotiaho J, Mappes J, Parri S, 1998. Mate choice for offspring performance: major benefits or minor costs? Proc R Soc Lond B 265: 2297-2301.
Andersson M, 1994. Sexual selection. Princeton, New Jersey: Princeton University Press.
Arak A, 1988. Female mate selection in the natterjack toad: active choice or passive attraction? Anim Behav 22: 317-327.
Baker MC, Bjerke TK, Lampe H, Espmark Y, 1986. Sexual response of female great tits to variation in size of male's song repertoires. Am Nat 128: 491-498.[Web of Science]
Barth FG, 1982. Spiders and vibratory signals: sensory reception and behavioral significance. In: Spider communication: mechanisms and ecological significance (Witt PN, Rovner JS, eds). Princeton, New Jersey: Princeton University Press; 67-122.
Baurect D, Barth FG, 1992. Vibratory communication in spiders I. Representation of male courtship signals by female vibration receptor. J Comp Physiol A 171: 231-243.
Brown WD, Wideman J, Andrade MCB, Mason AC, Gwynne DT, 1996. Female choice for an indicator of male size in the song of the black-horned tree cricket, Oeanthus nigricornis (Orthoptera: Gryllidae). Evolution 50: 2400-2411.[Web of Science]
Butlin RK, Hewitt GM, Webb SF, 1985. Sexual selection for intermediate optimum in Chorthippus brunneus (Orthoptera: Acrididae). Anim Behav 33: 1281-1292.
Catchpole CK, 1987. Bird song, sexual selection and female choice. Trends Evol Ecol 2: 94-97.
Catchpole CK, 1996. Song and female choice: good genes and big brains? Trends Evol Ecol 11: 358-360.
Cherry MI, 1993. Sexual selection in the raucous toad, Bufo rangeri. Anim Behav 45: 359-373.
Eiríksson T, 1993. Female preference for specific pulse duration of male songs in the grasshopper, Omocestus viridulus. Anim Behav 45: 471-477.
Eiríksson T, 1994. Song duration and female response behaviour in the grasshopper Omocestus virudulus. Anim Behav 47: 707-712.
Foelix RF, 1996. Biology of spiders. Oxford: Oxford University Press.
Forster L, 1982. Visual communication in jumping spiders (Salticidae). In: Spider communication. mechanisms and ecological significance (Witt PN, Rovner JS, eds). Princeton, New Jersey: Princeton University Press; 15-66.
Gerhardt HC, 1991. Female mate choice in treefrogs: static and dynamic criteria. Anim Behav 42: 615-635.[Web of Science]
Grafen, A. 1990. Biological signals as handicaps. J Theor Biol 144: 517-546.[Web of Science][Medline]
Gray DA, 1997. Female house crickets, Acheta domesticus, prefer the chirps of large males. Anim Behav 54: 1553-1562.[Web of Science][Medline]
Gray DA, Cade WH, 1999. Sex, death and genetic
variation: natural and sexual selection on cricket song. Proc R Soc
Lond B 266:
707-709.
Harrison JB, 1969. Acoustic behavior of a wolf spider, Lycosa gulosa. Anim Behav 17: 14-16.
Hasselquist D, Bensch S, von Schantz T, 1996. Correlation between male song repertoire, extra-pair paternity and offspring survival in the great reed warbler. Nature 381: 229-232.
Hedrick AV, 1986. Female preferences for male calling bout duration in a field cricket. Behav Ecol Sociobiol 19: 73-77.
Hergenröder R, Barth FG, 1983. The release of attack and escape behavior by vibratory stimuli in a wandering spider (Supiennus salei Keys.). J Comp Physiol 152: 347-358.
Hoikkala A, Aspi J, Suvanto L, 1998. Male courtship song frequency as an indicator of male genetic quality in an insect species, Drosophila montana. Proc R Soc Lond B 265: 503-508.[Medline]
Jang Y, Greenfield MD, 1996. Ultrasonic communication and sexual selection in wax moths: female choice based on energy and asynchrony of male signals. Anim Behav 51: 1095-1106.
Johnstone RA, 1995. Sexual selection, honest advertisement and the handicap principle: reviewing the evidence. Biol Rev 70: 1-56.[Medline]
Kotiaho JS, 2000. Testing the assumptions of conditional handicap theory: costs and condition dependence of a sexually selected trait. Behav Ecol Sociobiol 48: 188-194.[Web of Science]
Kotiaho JS, 2001. Costs of sexual traits: a mismatch between theoretical considerations and empirical evidence. Biol Rev 76: 365-376.[Medline]
Kotiaho JS, Alatalo RV, Mappes J, Nielsen MG, Parri S, Rivero A,
1998. Energetic costs of size and sexual signaling in a wolf
spider. Proc R Soc Lond B 265:
2203-2209.
Kotiaho J, Alatalo RV, Mappes J, Parri S, 1996. Sexual selection in a wolf spider: male drumming activity, body size, and viability. Evolution 50: 1977-1981.[Web of Science]
Kotiaho JS, Alatalo RV, Mappes J, Parri S, 1999. Sexual signalling and viability in a wolf spider (Hygrolycosa rubrofasciata): measurements under laboratory and field conditions. Behav Ecol Sociobiol 46: 123-128.
Kotiaho JS, Alatalo RV, Mappes J, Parri S, 2000. Microhabitat selection and audible sexual signaling in the wolf spider Hygrolycosa rubrofasciata (Araneae, Lycosidae). Acta Ethol 2: 123-128.
Kronestedt T, 1984. Ljudalstring hos vargspindeln Hygrolycosa rubrofasciata. Fauna Flora 79: 97-107.
Kronestedt T, 1996. Vibratory communication in the wolf spider Hygrolycosa rubrofasciata (Araneae, Lycosidae). Rev Suisse Zool (hórs serie) 341-354.
Lande R, 1981. Models of speciation by sexual
selection on polygenic characters. Proc Natl Acad Sci USA
78: 3721-3725.
Mappes J, Alatalo RV, Kotiaho J, Parri S, 1996.
Viability costs of condition-dependent sexual male display in a drumming wolf
spider. Proc R Soc Lond B 263:
785-789.
Møller AP, Alatalo RV, 1999. Good-genes effects
in sexual selection. Proc R Soc Lond B
266: 85-91.
Parri S, Alatalo RV, Kotiaho J, Mappes J, 1997. Female choice for male drumming in the wolf spider Hygrolycosa rubrofasciata. Anim Behav 53: 305-312.
Pomiankowski A, Møller AP, 1995. A resolution
to the lek paradox. Proc R Soc Lond
B:260:
21-29.
Prestwich KN, 1994. The energetics of acoustic signaling in anurans and insects. Am Zool 34: 625-643.
Reinhold K, Greenfield MD, Jang Y, Broce A, 1998. Energetic costs of sexual attractiveness: ultrasonic advertisement in wax moths. Anim Behav 55: 905-913.[Web of Science][Medline]
Rice WR, 1989. Analysing tables of statistical tests. Evolution 43: 223-225.[Web of Science]
Ritchie M, Townhill RM, Hoikkala A, 1998. Female preference for fly song: playback experiments confirm the targets of sexual selection. Anim Behav 56: 713-717.[Web of Science][Medline]
Rivero A, Alatalo RV, Kotiaho JS, Mappes J, Parri S, 2000. Acoustic signalling in a wolf spider: can signal characteristics predict male quality? Anim Behav 60: 187-194.[Web of Science][Medline]
Rovner JS, 1967. Acoustic communication in a Lycosid spider (Lycosa rabida Walckenaer). Anim Behav 15: 273-281.[Web of Science][Medline]
Rovner JS, Barth FG, 1981. Vibratory communication
through living plants by a tropical wandering spider. Science
214: 464-466.
Ryan MJ, 1988. Energy, calling, and selection. Am Zool 28: 885-898.
Ryan MJ, Keddy-Hector A, 1992. Directional patterns of female mate choice and the role of sensory biases. Am Nat 139: S4-S35.[Web of Science]
Schmitt A, Friedel T, Barth FG, 1993. Importance of pause between spider courtship vibrations and general problems using synthetic stimuli in behavioral studies. J Comp Physiol 172: 707-714.
Schmitt A, Schuster M, Barth FG, 1994. Vibratory communication in a wandering spider, Cupiennus getazi: female and male preferences for features of the conspecific male's releaser. Anim Behav 48: 1155-1171.
Schüch W, Barth FG, 1990. Vibratory communication in a spider: female responses to synthetic male vibrations. J Comp Physiol A 166: 817-826.
Searcy WA, Andersson M, 1986. Sexual selection and the evolution of song. Annu Rev Ecol Syst 17: 507-533.[Web of Science]
Simmons LW, 1988. The calling song of the field cricket, Gryllus bimaculatus (De Geer): constraints on transmission and its role in intermale competition and females choice. Anim Behav 36: 380-394.
Stratton GE, Uetz GW, 1981. Acoustic communication and
reproductive isolation in two species of wolf spiders. Science
214: 575-577.
Sullivan BK, Hinshaw SH, 1992. Female choice and selection on male calling behavior in the grey treefrog Hyla versicolor. Anim Behav 44: 733-744.
Uetz GW, Stratton GE, 1982. Acoustic communication and reproductive isolation in spiders. In: Spider communication: mechanisms and ecological significance (Witt P, Rovner JS, eds). Princeton, New Jersey: Princeton University Press; 123-159.
Vehrencamp SL, Bradbury JW, Gibson RM, 1989. The energetic cost of display in male sage grouse. Anim Behav 38: 885-896.[Web of Science]
Welch AM, Semlitsch RD, Gerhardt HC, 1998. Call
duration as an indicator of genetic quality in male gray tree frogs.
Science 280:
1928-1930.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  K. B. Karsten, L. N. Andriamandimbiarisoa, S. F. Fox, and C. J. Raxworthy Sexual selection on body size and secondary sexual characters in 2 closely related, sympatric chameleons in Madagascar Behav. Ecol., September 1, 2009; 20(5): 1079 - 1088. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. L Rodriguez, K. Ramaswamy, and R. B Cocroft Evidence that female preferences have shaped male signal evolution in a clade of specialized plant-feeding insects Proc R Soc B, October 22, 2006; 273(1601): 2585 - 2593. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. O. Elias, N. Lee, E. A. Hebets, and A. C. Mason Seismic signal production in a wolf spider: parallel versus serial multi-component signals J. Exp. Biol., March 15, 2006; 209(6): 1074 - 1084. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||