Behavioral Ecology Vol. 13 No. 3: 344-352
© 2002 International Society for Behavioral Ecology
Female calls in lek-mating birds: indirect mate choice, female competition for mates, or direct mate choice?
Department of Zoology, Norwegian University of Science and Technology, N-7491 Trondheim, Norway
Address correspondence to S.A. Sæther, who is now at the Department of Population Biology, Evolutionary Biology Centre, Uppsala University, Norbyvägen 18D, SE-752 36 Uppsala, Sweden. E-mail: stein.are.sather{at}chembio.ntnu.no .
Received 20 October 2000; revised 24 July 2001; accepted 24 July 2001.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
I tested predictions from ultimate hypotheses of why female great snipe Gallinago media give loud calls when visiting leks, using observational data and playback experiments. One hypothesis is that calls might be used in female—female competition for popular males, either (1) in an aggressive context toward a specific female, or (2) toward females in general to defend the male. It has also been suggested that female calls (3) may not have an adaptive function, the capability of vocalizing being explained as a correlated response to selection on male singing. Further, calls might function as (4) a copulation solicitation toward a specific male. Finally, calls might have a function in mate choice, either (5) in indirect mate choice as a fertility advertisement to incite male fighting, or (6) in direct mate choice as a mate-sampling aid to provoke quality-revealing responses from males. Disproportionately many female calls were uttered when no other females were present on a male's territory and in territories of males without mating success, contradicting hypotheses 1 and 2. Female calls were not associated with copulation; calls generally occurred several days before copulations, contradicting hypotheses 4 and 5. Playback of female calls attracted males and increased fighting among males, even if females were present nearby, contradicting hypothesis 3. Males changed their own songs in response to playback, and the response was related to their mating success. Taken together, the results are only consistent with one of the hypotheses considered—female calls may function as a mate-sampling aid used in direct mate choice.
Key words: female competition, female song, Gallinago media, great snipe, indirect mate choice, lekking, mate sampling.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Vocalizations during the breeding season, such as song, are common among males of most avian taxa. A multitude of investigations of male song has documented both inter- and intrasexual functions, such as territorial competition and maintenance and mate attraction (Catchpole and Slater, 1995
).
The function of female song is less well understood. Although males are
generally more vocal than females, females in a wide range of avian species
can produce song, and female song occurs in the wild in a number of species
(Langmore, 1998a,
2000), especially in the
tropics (Morton, 1996). Many
functional hypotheses for female song have been suggested. Recent work
indicates that female song is not an anomaly, but is instead used adaptively
in specific contexts (reviewed by Langmore,
1998a,
2000; see also
Eens and Pinxten, 1998;
Langmore, 1998b).
Loud female mating-season vocalizations occur in several lek-mating birds,
including the great snipe Gallinago media
(Sæther, 1994). Great
snipe females call when visiting leks, and the calls are different from those
given by males. Although one may not want to call this vocalization a
"song" (a phylogenetically biased concept), its occurrence merits
the question of what its adaptive function, if any, might be. At least five
ultimate hypotheses may explain female song in lekking birds (outlined below).
The aim of this study was to test these hypotheses, which are not necessarily
mutually exclusive. I also considered the hypothesis that female song has no
adaptive function. Here I also briefly discuss some additional, but unlikely,
explanations. Due to the nature of the lek mating system, some hypotheses
suggested for other birds may be readily dismissed as not relevant to female
vocalizations at leks and are not considered further here. This includes
coordination of breeding activities, territorial defense, attraction of
extrapair copulations, mate attraction, and locating fledglings (Langmore,
1998a,
2000).
Hypotheses
The female—female competition hypotheses
Female signing has been found to be associated with female—female
competition in several recent studies (Langmore,
1998a,
2000). However, in lekking
species, females are generally not thought to be reproductively limited by
access to males because ready-to-mate males are plentiful. Yet intrasexual
aggressive behavior by females has been reported from some studies
(Karvonen et al., 2000;
Petrie et al., 1992;
Sæther et al., 2001).
This might be expected because of competition for ejaculates of widely
preferred males (Petrie et al.,
1992). If calls are used in female—female competition, I
predicted that loud calls by females should occur on the territories of
successful males (i.e., those males that obtained matings). It cannot be
predicted from this hypothesis that female calls should necessarily be given
in agonistic contexts or that females should respond aggressively toward
playbacks because calls could be facilitatory rather than agonistic in nature.
Therefore, I split this hypothesis into two distinct subhypotheses. (1) Female
calls may be an aggressive signal directed at a specific female that may
otherwise mate with the calling female's preferred male. This first hypothesis
then makes the additional prediction that calls should only occur when more
than one female is present on a male territory. (2) Females call to discourage
other females from attempting to mate with the male; hence calls may be
uttered from solitary females but should still occur on successful males'
territories.
The no-adaptive-function hypothesis
It could be argued that female song has no adaptive significance (e.g.,
Nice, 1943;
Thorpe, 1964), a possibility
that should always be considered for any trait studied by adaptationists
(e.g., Dennett, 1995). From
population genetics theory, a nonadaptive trait can be expected to be present
in members of one sex if it bears no cost, as a corollary to selection for the
trait in the other sex (Halliday and
Arnold, 1987; Muma and
Weatherhead, 1989). Nice
(1943) suggested that, for
species in which female song is rare (e.g., the song sparrow, Melospiza
melodica), the singing females are abnormal individuals showing unusually
high androgen levels and female song represents a vestigial phenomenon. Arcese
et al. (1988) concluded that
female song may be anomalous and therefore unimportant in such species. If the
female vocalization has no function in great snipe, I predicted that the birds
should show no response to the call, beyond the effect of female presence per
se.
The copulation solicitation hypothesis
The female call might simply indicate a readiness to mate (e.g.,
Hannon, 1980), directed to a
specific male. If this is so, I predicted that calls should be associated with
copulations and occur on territories of males that obtain matings.
Indirect mate choice: the fertility advertisement hypothesis
Montgomerie and Thornhill
(1989) proposed that female
vocalization during the breeding season is an advertisement of fertility that
incites male—male competition over access to the female (see also
Thornhill, 1988), thus
improving the probability that a female will mate with a fit male. Cox and Le
Boeuf (1977) presented
essentially the same argument. Wiley and Poston
(1996) used this to illustrate
the concept of indirect mate choice (i.e., female behavior that restricts her
set of potential mates by biasing the outcome of male competition for mates).
A comparative study (Montgomerie and
Thornhill, 1989) showed that, as predicted by the hypothesis, loud
calls made by females during the breeding season have been disproportionally
often reported in birds with multimale mating systems, such as lekking. The
fertility advertisement hypothesis might also apply to socially monogamous
species if female vocalizations attract neighboring males (e.g., bearded tit
Panurus biarmicus; Hoi,
1997). If females advertise their fertility by giving loud calls
and thereby incite male—male competition, I predicted that calls should
be associated with copulation and that the frequency of fighting among the
males should increase after a female call.
Direct mate choice: the mate-sampling aid hypothesis
Females might also use vocalizations in direct mate choice as a means of
sampling males, by provoking a response from males and evaluating their
performance. In contrast to the hypothesis above, this does not imply that an
advertisement of fertility is involved, or that male—male fighting is
necessarily important. If this idea is correct, loud calls by females should
occur in the mate-sampling period of females (i.e., early in the breeding
season, before mating has taken place). Further, I predicted that males should
respond to the call, and, on condition that the relevant variable is measured,
the responses should differ in relation to males' mating success.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Study species and general methods
The great snipe is an almost monomorphic species (Höglund et al., 1990
) of
Eurasian shorebird that exhibits classical lek breeding
(Höglund and Lundberg,
1987; Lemnell,
1978). Males are active on the lek arena during the night-time in
spring, where they perform stereotyped displays on the ground, including a
vocalization (Figure 1b). Males
defend adjacent territories of about 100 m2
(Höglund and Lundberg,
1987). My study area is situated in the subalpine and low-alpine
regions, from 1000 to about 1200 m altitude, in Gåvålia near
Kongsvoll (62°17' N, 9°36' E), Dovrefjell, central Norway. About five
leks are occupied each season in the study area, and about 30 males may defend
territories at the largest lek.
|
This study is primarily based on data from lek 2
("Hestesletta") from 1992 to 1994, but additional data on
identified females observed calling are included for other leks from 1988 to
1998. Data refer solely to lek 2 unless otherwise stated. Birds were captured
(mostly in mid- to late May) using mist nets at the lek sites. Each bird was
uniquely marked with colored leg-rings and a numbered steel bird-ring. I
addition, most of the females were given a combination of strips of colored
adhesive tape on their backs to ease field identification. Birds were sexed
according to bill length (Höglund et
al., 1990), and most were aged according to primary feather wear
(Sæther et al., 1994).
My colleagues and I made field observations from elevated hides (1.75 m above
ground) erected at the leks. Great snipes gather at night, and torches often
had to be used to see the birds, but this did not affect their behavior
(Sæther, unpublished data). Lek 2 was visited by one to three observers
on most dates between 15 May and 15 June in 1992-1994 (see
Figure 2). At this lek, 30
males were observed in 1992, including 22 resident males (observed on at least
five nights). In 1993 the corresponding numbers were 27 and 23, and in 1994,
33 and 19. Each observer monitored behavior of males and females on the
territories of four to eight males each night, from 2300 h in the evening
until 0300 h in the morning (daylight savings time). Dates refer to the date
of the start (the evening) of an observational period.
|
Males that copulated at least once in a given year were classified as
successful in that year, whereas those that were not observed copulating were
classified as unsuccessful. I assigned copulatory success only to males that
were present at the lek for at least 5 days on which field observations were
made. Because individual females often mate several times during the mating
season (Fiske and Kålås,
1995), I did not simply sum the total number of copulations a male
received as a measure of mating success. Instead, I tried to estimate the
minimum number of individual females that mated with the male. This was done
by combining the records for identified females with the minimum number of
different unmarked females mating with the male. This latter number was
calculated by the criteria that an unmarked female could have mated with the
male on a maximum of 3 consecutive nights
(Fiske and Kålås,
1995). I could sometimes keep track of several unmarked females
and thus separate between different unmarked females mating on the same night,
but not across nights. Marked females accounted for 22 of the 35 females
estimated to have copulated at lek 2 in 1992. The corresponding numbers were 5
out of 16 females in 1993 and 19 out of 24 females in 1994.
Observations of loud calls by females
Females utter loud calls (Figure
1a) while on the lek
(Sæther, 1994). When
visiting leks, females usually either sit or move around with their bodies
held low. When females utter the loud call, they stand up in a male-like
manner with their breasts raised, but normally they do not spread out their
white tails or raise their heads as males do when vocalizing (see
Lemnell, 1978, for a
description of male display behavior). After calling, females quickly resume
an inconspicuous appearance. The calling behavior is thus a rather conspicuous
event among otherwise secretive females. At lek 2, I noted 63, 50, and 111
loud female calls in 1992, 1993, and 1994, respectively (see
Sæther, 1994).
For each female loud call heard, the following information was obtained when possible: female identity, which male territory she was in, time of day, male reaction, reaction of other females, number of females present on the territory, whether copulation followed within 5 min, and if the female was involved in any agonistic interactions with other females when she vocalized. Agonistic interactions may take the form of one female quickly running against a nearby female, who then withdraws, or one or more females walking or running in a male-like manner with their tails spread out and raised toward their backs, in the presence of other females. For some loud calls, it was impossible to establish which female made the sound, and in such cases only the time of day was recorded. Cases in which the vocalizing female was located are referred to as "seen calls," which also includes unmarked females.
Analysis of number of female calls in relation to male mating success
and number of females on the territory
Because successful males may have more females present on their territories
than do unsuccessful males, it might be expected that successful males should
have more female loud calls even if females did not preferentially call at
such males' territories. Hence, I compared the observed proportions of female
calls on territories of successful and unsuccessful males against the
proportions expected by chance. These expected proportions were calculated
from the number of females present at territories of successful and
unsuccessful males each time a loud call by a female was observed. These data
were obtained from all records of female movements each night, allowing me to
calculate the exact number of females present at each male territory whenever
a loud call was observed.
Analogous to the above, I compared the observed proportions of female calls occurring when the female was the sole female on the territory against the expected proportions. These expected proportions were calculated from the number of solitary and accompanied females present at the lek each time a loud call by a female was observed.
Playbacks of female calls
I recorded male vocalizations before and after playback of a female loud
call to quantify differences in response between males with mating success and
those without. This was initially done on 19, 24, and 25 June 1992 at lek 2.
The female call was recorded on 6 June 1991 (identical to the one reproduced
in Figure 1a). For each male, I
broadcasted the female call once. I made recordings using a Sony TCD-5 Pro
tape recorder and an AKG ck9 directional microphone before and after the
playback. The playback was broadcasted using loudspeakers connected to a
cassette player operated from inside a blind. In advance, I adjusted the
output volume to be close to that of a natural female loud call. The
loudspeaker was situated on the territorial boundary between two males, and
vocalizations of both males were recorded before and after playback. Only one
recording of each male before and one after playback was used in the
subsequent analyses. I tried to use the last display made before and the first
one after playback; this was generally within 1 min of the playback. If a male
had several neighboring territories, he might have been exposed to a playback
more than once; I used the recordings from the first exposure unless later
recordings were closer in time to a playback. No females were observed on the
lek on the days of the experimental playbacks in 1992.
To see if the results obtained in 1992 were robust, I replicated the playback experiments in 1993. This was done at both lek 2 and at lek 5 on 22, 24, and 30 May and 4, 5, and 8 June, using the same procedure as in 1992. Three males that appeared in the sample during both years were included only from 1992 in the pooled sample. Recordings from 10 males with mating success and 15 without were used in the analyses (6 successful and 7 unsuccessful in 1992, and 4 successful and 8 unsuccessful in 1993).
Playbacks were also carried out at lek 10 in 1992 early in the season when females were still present. This was done as a pilot study to see if playback of a female call had any effect on males and females. The loudspeaker was placed on the territorial boundary between two males. I noted if males were immediately attracted (i.e., within a few seconds after playback) and if the frequency of fighting increased.
Potential problems of pseudoreplication by using the same recording in all playbacks could not be overcome because only one recording of a female loud call was available.
Analysis of male response in song
I chose to analyze one measure of frequency (Hz, last melodic note; see
Figure 1b) and one measure of
duration (seconds, strophe length; Figure
1b). This measure of strophe length was chosen because it was the
most accurate and unambiguous measure of duration, and the measure of
frequency was chosen as the one with highest frequency that could accurately
be measured. The analyses were made using a Kay DSP 5500 Sonagraph. Frequency
and strophe length were measured in narrow and wide band modes, respectively.
Frequency was measured to the nearest 0.01 kHz, and strophe length was
measured to the nearest 0.001 s. I tried to control for date in the
statistical analyses because song characteristics might change during the
season. I also took male age into consideration because age might confound
relationships between success and male traits in this species
(Fiske et al., 1994).
Statistical analyses were made with the SPSS 4 program for the Macintosh computer. Probability values are two-tailed. Nonparametric tests were corrected for ties.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Do calls occur most often on preferred males' territories?
Significantly more loud calls by females were observed on unsuccessful males' territories than expected from the proportion of observed females on unsuccessful males' territories both in 1992 and 1994 (Figure 3). In 1993 all calls and all observed females at the time of calling were on successful males territories, preventing statistical testing. Significantly more female calls than expected were observed on unsuccessful males territories when the data from all years were pooled (Figure 3).
|
Do calls occur most often when more than one female is present in the
territory?
More females than expected made loud calls when alone both in 1992 and 1994
(Figure 4). In 1993, all female
calls and all females observed at the time of calling were solitary. In the
pooled data set for all years, significantly more solitary females gave calls
than expected (Figure 4).
|
Eight of 50 seen calls in 1992 (on three leks) were made during female—female agonistic interactions. One of these instances involved the chasing of another female by the vocalizing female. In 1993 no calls were given in agonistic contexts. Only one of 51 seen calls in 1994 was made in a female—female agonistic interaction.
Are calls associated with copulation?
The numbers of loud calls by females (seen calls on 3 leks) followed by
copulation by the same female within 5 min were 2 out of 8 in 1991 and 1 out
of 50 in 1992. In 1993 and 1994 no such cases were recorded, but copulation
after female calls has sporadically been observed since. On lek 2 in 1992,
only 1 of 36 seen calls was followed by copulation, and only 1 of 48
copulations was preceded by a loud call.
Do calls predominantly occur in the premating period?
Loud calls generally occurred earlier in the season than did copulations
both in 1992 (Mann-Whitney U test, z = -6.63, p
<.0001; Figure 2a) and in
1994 (z = -10.09, p <.0001;
Figure 2c) but not in 1993
(z = -1.17, p =.24;
Figure 2b).
Data from 18 females observed both to vocalize and copulate in the same year, 1988-1998, showed that calls occurred on average 3.4 days before a female's first copulation, although two females had copulated before they called, and two copulated later the same day. The majority (78%) vocalized when they were in the territory of the male they also mated with.
Indications that most of the females making loud calls did so early in their lek-visiting period stem also from the large proportion of unmarked females vocalizing. Females that made loud calls were less likely to have been previously caught and marked than expected from the observed proportion of marked females throughout the season (data from 1992, summed totals for all days, 3 of 19 calling, 79 of 131 observed, binomial p <.0001). Most vocalizing females of known age were older than 1 year (87%, n = 23, mean = 3.7 years), but this might be because many first-year females had not yet been marked when vocalizing.
Do calls increase male fighting?
Playback of female loud calls at lek 10 in 1992 led invariably to an
immediate and obvious attraction of males toward the source of the call. It
also led to an obvious increase in fighting among males. No fighting occurred
immediately before playback, but in 8 of 11 trials nearby males started
fighting within seconds after a playback. During all of these trials females
were present at the lek. In the playback experiments described below, females
were not always present, but males were also then invariably attracted to the
calls, and most often fighting was obviously triggered by the playback.
Do successful and unsuccessful males respond differentially to female
calls?
Before playback, successful and unsuccessful males had similar strophe
lengths (Figure 5).
Unsuccessful males increased their strophe length after the playback
(comparing pre- and post-playback strophe length, paired t test,
t = -3.09, df = 14, p =.008;
Figure 5). In contrast,
successful males did not change their strophe length in response to playback
of female calls (paired t test, t = 1.38, df = 9, p
=.2; Figure 5). The
strophe-length response (change from pre- to post-playback) differed
significantly between successful and unsuccessful males (t test,
t = -3.03, df = 23, p =.006).
|
I found a significant negative correlation between the minimum number of females a male copulated with and the response in strophe length (rs = -.53, n = 25, p =.006), even after controlling for male age (ranked values, rpartial = -.51, n = 23, p =.015) and date (ranked values, rpartial = -.54, n = 25, p =.007).
For the frequency (Hz) of the last melodic note, a significant response was found (paired t test, t = -2.09, df = 24, p =.047; Figure 6), but the response did not differ according to success (t test, t = -.51, df = 23, p =.62; Figure 6).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Below I discuss the evidence for and against the different hypotheses, and briefly mention some other unlikely explanations. I conclude that the mate-sampling aid hypothesis is the only explanation supported of those considered. I discuss how females in this and other species may use vocalizations in their mate-sampling process without signaling fertility, and how this relates to the concepts of direct and indirect mate choice.
Competition among females for preferred males
Contrary to both versions of the female—female competition
hypothesis, I found that calls occurred less often on successful males'
territories than expected and less often when other females were present than
expected. A few loud calls were uttered in agonistic contexts. Because so many
calls were observed at territories of males that never obtained matings, this
study does not support the female competition hypothesis as a general
explanation for the function of the loud call of female great snipes.
Conflicts of interest between great snipe females over access to ejaculates of
popular males may exist (Sæther et
al., 2001), but female calls do not seem to have a function in
this context. It is possible, however, that females may use the call for
multiple purposes.
No adaptive function
Contrary to the no adaptive function hypothesis, playbacks of female loud
calls attracted males and led to increased fighting among males. Males
responded even if females were present nearby at the lek. Males seem to be
well aware of the presence of silent females even on neighboring territories,
as indicated, for example, by moving closer to them (Sæther, personal
observations). Thus, the males seemed to respond to the loud call per se,
rather than to mere female presence. However, a more robust test is needed
before being confident that male responses to playbacks are responses to
female calls because no attempt was made to compare with the response to a
control sound. There were no indications whatsoever that females responded to
naturally occurring calls. Unfortunately, this cannot be analyzed properly
from the playback experiments because few females were present, but those that
were did not move toward the speaker.
Other observations against calls having no adaptive value is that calls only occur at leks, in a very restricted time period, and draw attention toward females that otherwise are inconspicuous. These observations indicate that calling entails costs, and hence a benefit is expected.
Copulation solicitation
The female loud call does not seem to be a solicitation behavior (a signal
of readiness to copulate with a specific male). Calls were only rarely
associated with copulation or solicitation and occurred on average several
days ahead of copulations. When females solicit copulation they sometimes
utter a faint sound, which is distinct from the loud call
(Sæther, 1994).
Fertility advertisement to incite fighting
Contrary to predictions from the fertility advertisement hypothesis, as
discussed above, calls were rarely followed by immediate copulation and
occurred, on average, several days before copulation. The data on individual
females observed both vocalizing and copulating also support the conclusion
that females vocalize several days before copulating. Fertility advertisement
is hence an unlikely explanation for why females give loud calls in this
species. Other evidence suggests that male access to mating with females is
not directly determined by male interactions, as this hypothesis assumes.
Sæther et al. (1999)
found that males almost exclusively copulate within their own territory, and
matings are always preceded by a female solicitation. The females do not
passively copulate with the male winning a fight as Montgomerie and Thornhill
(1989) implicitly suggested,
and as Wiley and Poston (1996)
used as an example of indirect mate choice.
I suspect that some cases where females of other species have been claimed
to advertise fertility and incite fighting to secure mating with a fit male
may also prove to be unjustified. Calls associated with copulation may instead
be explained by direct benefits to the female of reducing harassment from
pushy subordinate males by evoking the attention of a dominant male
(Clutton-Brock and Parker,
1995). Calling need not have much to do with mate choice. Cassini
(1999) similarly suggested that
female calls in Northern elephant seals Mirounga angustirostris are
better understood as a means of harassment avoidance rather than a fertility
announcement as originally proposed by Cox and Le Boeuf
(1977). Further, female calls
of the fowl Gallus gallus suggested by Thornhill
(1988) to be fertility
announcements have now been found by Pizzari and Birkhead
(2001) to be neither
associated with copulation nor with increased fighting among males. Instead,
female fowl use another call to trigger dominant males to disrupt copulation
attempts by low-ranking males when other methods of avoiding copulation have
failed (Pizzari, 2001). The
temporal association of calls with fertility and fighting in some species
(Montgomerie and Thornhill,
1989; Wiley and Poston,
1996) might be because harassment is then more costly to females
(Cassini, 1999) or simply
because males are more eager to harass females at this time
(Sæther et al.,
1999).
We thus face a minor paradox: in species where the fertility announcement hypothesis was originally proposed because of an association with copulation, calls might give direct benefits to females and need not be involved in (indirect) mate choice based on fighting at all. However, in species such as the great snipe where calls are not associated with copulation, calls may indeed have a function in (direct) mate choice, and it cannot be ruled out that fighting performance is an important cue for females (see below).
Other unlikely explanations
Cheng (1992) reviewed
studies of female nest cooing in ring doves Streptopelia risoria that
suggest a self-stimulation function for the female call. There is evidence in
ring doves that the female's own calls affect her endocrine state. This
hypothesis is consistent with my observation that female calls generally
occurred well before copulation, if we make the assumption that a female calls
to change her own endocrine state so as to become fertile. However, the
hypothesis is not consistent with the observation that the males seemed to
respond to the female call per se. It is also unclear why the females should
vocalize when at the lek if they are stimulating their own endocrine states.
Great snipe females are not known to utter the loud call anywhere else but at
leks (Sæther, 1994).
Another unlikely idea is that calls have evolved because of male mate
choice. An element of mutual mate choice exists in great snipes
(Sæther et al., 2001),
but it is hard to see how any benefits of male choosiness based on assessment
of female calls (variation in quality among females signaled by the call)
could exceed the costs (reduced reproductive rate due to lost matings). The
only possibility seems to be that, because popular males often refuse to
remate with a female if unmated females are present at the territory
(Sæther et al., 2001),
females might call to signal that they have not yet copulated. Notwithstanding
the problem of how honesty can be maintained, and regardless of what quality
is signaled, we would expect calls to be associated with copulation and only
directed to males achieving mating success. Both these predictions are refuted
by the data.
Another explanation is that females may be trying to attract other females, perhaps because it is risky being a lone female on a lek. This seems unlikely for several reasons. First, giving a call is a conspicuous behavior among otherwise secretive females, meaning that calling may attract predators. I have never observed a predator attempting to catch a female at a lek, but the much more conspicuous males are sometimes caught. Second, no female responses could be noticed, either to naturally occurring calls or playbacks. Third, most female calls did in fact occur when the female was accompanied by other females (often several), even if this occurred less often than expected from the distribution of females (Figure 4).
In a lekking species where the female call is associated with copulation,
eavesdropping females might use calls as a cue for mate-choice copying (as
suggested by Bradbury and Gibson,
1983, for the hammer-headed bat, Hypsignathus
monstrosus). A possibility is therefore that the function of the female
call is to initiate copying, if there is any benefit attained to being copied.
Such benefits might possibly exist if copying changes the mean preference of
traits in the population (in the present or subsequent seasons) and if a
signaler's (i.e., a calling female's) kin benefit more than the kin of average
females. However, there may also be costs associated with being copied.
Evidence from both this species
(Sæther et al., 2001)
and for other lekking birds (Petrie et
al., 1992) suggest that females compete for preferred males at a
lek. The initiate-copying idea is refuted by my observations because female
calls did not predominately occur on territories of successful males, and
calls were not associated with copulations. Further, Fiske et al.
(1996) found that the best
estimate of the proportion of copying females is zero in great snipe.
Mate sampling aid
The mate-sampling aid hypothesis received some support from the finding
that unsuccessful and successful males differed in their response to playbacks
(discussed below). Since the same female's call had to be used in all
playbacks, these results should not be interpreted as a robust test of
differential response, but only as preliminary supporting evidence for the
hypothesis. Further support was that loud calls occurred in the potential
mate-sampling period of females, several days before copulations. Fiske and
Kålås (1995)
showed that great snipe females spend several days on the lek before
copulating and that most females visit several male territories before
copulating. A few females were observed to call also later than their first
copulation, but as females usually copulate on more than one night, they may
continue mate sampling after copulation commences. The results from 1993
(Figure 2b) suggest that in
this year loud calls did not occur earlier than copulations. This is likely to
be because much mate sampling (and presumably calling) took place early that
year, before observers were present at the lek, as was reflected in the early
occurrence of copulations and the few calls heard.
Some of the other results may also be explained by this hypothesis. Let us
assume that females can also make use of male responses to calls of other
females present. This may explain why more calls than expected were observed
when females were alone on male territories. Let us also assume that females
have a given probability of either vocalizing themselves or using another
female's call once at each male territory they visit and that females spend
less time with males they do not copulate with
(Fiske and Kålås,
1995; Sæther et al.,
1999). Then we should find, as was observed, more female calls
occurring in unsuccessful males' territories than expected from the
distribution of females.
For many are called, but few are chosen: how may females use calls as
an aid in mate choice?
If females use loud calls as a mate sampling aid to test male quality, what
traits should males display as a response? One candidate is coverable
ornaments. This is unlikely to be important in great snipes because the only
such trait is the male's white tail, which is displayed unsolicited at a rate
of about 2.5 displays/min simultaneously with the male vocalization
(Fiske et al., 1994). Another
possibility is a change in the frequency of repeated behavior such as
displays. It seems likely, however, that females would gain more honest
information from paying attention to the long-term display rate rather than
using calls to incite a temporal increase, particularly if a female sample
males over long periods as in great snipes
(Fiske and Kålås,
1995). Male display rate appears to be unrelated to mating success
in this species (Fiske et al.,
1994; but see Höglund and
Lundberg, 1987).
Fighting is another behavioral change that might be brought on by female
calls. In great snipes, male fights do not result in immediate access to
females; no forced copulations have been observed despite many thousand hours
of observation on leks. When a female is disturbed on a male's territory by
neighboring males, she flies away and often returns to the same territory
within a few minutes (Sæther et al.,
1999). Nevertheless, it is possible that a female may evaluate the
males on the basis of their performance in combat during her mate sampling
activity, and she may indeed use her calls to incite fighting. This may
explain why great snipe males responded to female calls by fighting, though
without gaining any immediate benefit (i.e., copulation). This highlights the
important difference between the mate-sampling aid hypothesis and the
fertility advertisement hypothesis: direct versus indirect mate choice. Wiley
and Poston (1996) clarified
the concepts of mating competition and mate choice by introducing the concept
of indirect mate choice. Indirect mate choice occurs when females behave such
as to bias the outcome of male competition in favor of certain males, but
without exerting any discrimination of males per se. However, females might
use male performance in fighting as a cue in direct mate choice. This suggests
that the commonly assumed clear-cut distinction between intra- and intersexual
selection may be even more ambiguous than Wiley and Poston
(1996) recognize (but see
Fisher, 1958). Many traits
used as armaments in male interactions seem also to be favored as ornaments by
direct female preferences (Berglund et al.,
1996; see also Sæther et
al., 1999 for a discussion of the concepts). However, females do
not prefer dominant males in this species
(Sæther et al., 1999),
and benefits of mating with dominant males may generally have been
overemphasized (Qvarnström and
Forsgren, 1998).
Another cue that females might assess during their mate sampling is male
vocal characteristics. Höglund and Lundberg
(1987) presented some evidence
that properties of the males' display call are related to mating success in
great snipes. If females use their loud call to evaluate such traits, it might
be expected that male response to female loud calls should be related to male
mating success. Indeed, this is what was found in the playback experiments.
However, it is unclear why the unsuccessful males increased their strophe
length in response to the female call, while the successful males did not. I
suspect that females assess cues linked to strophe length that the
unsuccessful males fail to perform without also increasing their strophe
length. This might be related to the frequency level of the song because males
increased somewhat the frequency of the last melodic note in response to
playback regardless of their mating success.
Does the mate-sampling aid hypothesis apply to female breeding-season
vocalizations in other species?
I suggest that the females of great snipe and of other lekking species that
utter loud calls during the breeding season, such as sharp-tailed grouse
Tympanuchus cupido (Montgomerie
and Thornhill, 1989) and black grouse Tetrao tetrix
(Gerber, 1955; Höglund J,
personal communication), use these calls to increase the accuracy of their
mate choice. This need not involve a fertility advertisement. In the
promiscuous blue grouse Dendragapus obscurus, Hannon
(1980) found that the females
started to cackle several days before copulation commenced and that the males
responded to playbacks by increasing their rate of "whooting" (a
courtship call). Hannon considered three adaptive explanations for the female
cackle: a signal of readiness to mate, a means of synchronizing the breeding
cycles of males and females, and as an aggressive signal to other females. The
first two explanations were rejected because, although males were attracted to
females that cackled, these females were not necessarily receptive and because
the males already were in breeding condition when females started to cackle.
The aggression hypothesis received some support. These findings, however, are
consistent with my mate-sampling aid hypothesis because the females cackled
during their potential mate-sampling period and the males responded by
increasing their display rate (Hannon,
1980). Hannon's results are not consistent with the fertility
advertisement hypothesis because cackling did not lead to copulation.
If performance in male—male interactions is not important as a mate choice cue, then females may conceivably use calls to evaluate males one by one. Thus, the mate-sampling aid hypothesis might not only apply to lekking and other mating systems where multiple males are present in a restricted space, but it may be of more general interest for understanding the evolution of female vocalizations.
|
ACKNOWLEDGEMENTS |
|---|
I thank T. Bretten, P. Fiske, J.M. Gjul, J.A. Kålås, S. Kålås, I. Myklebust, T.H. Ringsby, and S.L. Svartaas for excellent help and company in the field, and T. Amundsen, I. Cuthill, Y. Espmark, P. Fiske, J. Höglund, G. Högstedt, J.A. Kålås, A. Lundberg, T. Opdahl, P.T. Rintamäki, P.T. Smiseth, F. Widemo, and reviewers for valuable discussions or comments on earlier drafts.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Arcese P, Stoddard PK, Hiebert SM, 1988. The form and function of song in female song sparrows. Condor 90: 44-50.[Web of Science]
Berglund A, Bisazza A, Pilastro A, 1996. Armaments and ornaments: an evolutionary explanation of traits of dual utility. Biol J Linn Soc 58: 385-399.
Bradbury JW, Gibson RM, 1983. Leks and mate choice. In: Mate choice (Bateson P, ed). Cambridge: Cambridge University Press; 109-138.
Cassini MH, 1999. The evolution of reproductive
systems in pinnipeds. Behav Ecol 10:
612-616.
Catchpole CK, Slater PJB, 1995. Bird song: biological themes and variations. Cambridge: Cambridge University Press.
Cheng M-F, 1992. For whom does the female dove coo? A case for the role of vocal self-stimulation. Anim Behav 43: 1035-1044.
Clutton-Brock TH, Parker GA, 1995. Sexual coercion in animal societies. Anim Behav 49: 1345-1365.
Cox CR, Le Boeuf BJ, 1977. Female incitation of male competition: a mechanism in sexual selection. Am Nat 111: 317-335.[Web of Science]
Dennett DC, 1995. Darwin's dangerous idea: evolution and the meanings of life. New York: Simon and Schuster.
Eens M, Pinxten R, 1998. Female song for mate attraction: an over-looked phenomenon? Trends Ecol Evol 13: 322-323.
Fisher RA, 1958. The genetical theory of natural selection, 2nd ed. New York: Dover.
Fiske P, Kålås JA, 1995. Mate sampling and copulation behavior of great snipe females. Anim Behav 49: 209-219.
Fiske P, Kålås JA, Sæther SA, 1994.
Correlates of male mating success in the lekking great snipe Gallinago
media: results from a four-year study. Behav Ecol
5: 210-218.
Fiske P, Kålås JA, Sæther SA, 1996. Do female great snipe copy each others mate choice? Anim Behav 51: 1355-1362.
Gerber R, 1955. Singende vogelweibchen. Beitr Vogelkd 5: 36-45.
Halliday T, Arnold SJ, 1987. Multiple mating by females: a perspective from quantitative genetics. Anim Behav 35: 939-941.
Hannon SJ, 1980. The cackle call of female blue grouse: does it have a mating or aggressive function? Auk 97: 404-407.[Web of Science]
Höglund J, Lundberg A, 1987. Sexual selection in a monomorphic lek-breeding bird: correlates of male mating success in the great snipe Gallinago media. Behav Ecol Sociobiol 21: 211-216.
Höglund J, Kålås JA, Løfaldli L, 1990. Sexual dimorphism in the lekking great snipe. Ornis Scand 21: 1-6.
Hoi H, 1997. Assessment of the quality of copulation partners in the monogamous bearded tit. Anim Behav 53: 277-286.[Web of Science]
Karvonen E, Rintamäki PT, Alatalo RV, 2000. Female-female aggression and female mate choice on black grouse leks. Anim Behav 59: 981-987.[Web of Science][Medline]
Langmore NE, 1998a. Functions of duet and solo songs of female birds. Trends Ecol Evol 13: 136-140.
Langmore NE, 1998b. Reply from N.E. Langmore. Trends Ecol Evol 13: 323.
Langmore NE, 2000. Why female birds sing. In: Animal signals: signalling and signal design in animal communication (Espmark Y, Amundsen T, Rosenqvist G, eds). Trondheim: Tapir; 317-327.
Lemnell PA, 1978. Social behaviour of the Great Snipe Capella media at the arena display. Ornis Scand 9: 146-163.
Montgomerie R, Thornhill R, 1989. Fertility advertisement in birds: a means of inciting male-male competition? Ethology 81: 209-220.[Web of Science]
Morton ES, 1996. A comparison of vocal behaviour among tropical and temperate passerine birds. In: Ecology and evolution of acoustic communication in birds, (Kroodsma DE, Miller EH, eds). Ithaca, New York: Cornell University Press; 258-268.
Muma KE, Weatherhead PJ, 1989. Male traits expressed in females: direct or indirect sexual selection? Behav Ecol Sociobiol 25: 23-31.[Web of Science]
Nice MM, 1943. Studies in the life history of the song sparrow. II. Trans Linn Soc NY 6: 1-328.
Petrie M, Hall M, Halliday T, Budgey H, Pierpoint C, 1992. Multiple mating in a lekking bird: why do peahens mate with more than one male and with the same male more than once? Behav Ecol Sociobiol 31: 349-358.[Web of Science]
Pizzari T, 2001. Indirect partner choice through manipulation of male behaviour by female fowl, Gallus gallus domesticus. Proc R Soc Lond B 268: 181-186.[Medline]
Pizzari T, Birkhead, TR, 2001. For whom does the hen cackle? The function of postoviposition cackling. Anim Behav 61: 601-607.
Qvarnström A, Forsgren E, 1998. Should females prefer dominant males? Trends Ecol Evol 13: 498-501.
Sæther SA, 1994. Vocalizations of female great snipe Gallinago media at the lek. Ornis Fenn 71: 11-16.
Sæther SA, Kålås JA, Fiske P, 1994. Age determination of breeding shorebirds: quantification of feather wear in the lekking great snipe. Condor 96: 959-972.[Web of Science]
Sæther SA, Fiske P, Kålås JA, 1999.
Pushy males and choosy females: courtship disruption and mate choice in the
lekking great snipe. Proc R Soc Lond B
266: 1227-1234.
Sæther, SA, Fiske P, Kålås JA, 2001. Male mate choice, sexual conflict and strategic allocation of copulations in a lekking bird. Proc R Soc Lond B 268: 2097-2102.[Medline]
Thornhill R, 1988. The jungle fowl hen's cackle incites male competition. Verh Dtsch Zool Ges 81: 145-154.
Thorpe WH, 1964. Singing. In: A new dictionary of birds (Thomson AL, ed). London: Thomas Nelson & Sons; 739-750.
Wiley RH, Poston J, 1996. Indirect mate choice, competition for mates, and coevolution of the sexes. Evolution 50: 1371-1381.[Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||