| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Behavioral Ecology Vol. 14 No. 4: 521-525
© 2003 International Society for Behavioral Ecology
No peace for estrous topi cows on leks
Institute of Zoology, Zoological Society of London, Regent's Park, London NW1 4RY, UK
Address correspondence to J. Bro-Jørgensen, who is now at the Department of Biological Sciences, University of Stirling, Stirling FK9 4LA, UK. E-mail: jb24{at}stir.ac.uk.
Received 5 March 2002; revised 12 September 2002; accepted 1 October 2002.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Male coercion, such as harassment, may be considered the third main component in sexual selection alongside male competition and female choice. In this study on lek-breeding topi antelopes (Damaliscus lunatus), I investigate whether female mating preferences have consequences for male investment in harassment and whether harassing males are more likely to succeed in mating. I then address the question of whether lek evolution in topi can be explained by harassment avoidance. Judging from mating rate, I found that female topi antelopes in estrus preferred lek males to resource defenders. In contrast to lek males, resource defenders demonstrated significantly higher harassment rates before they succeeded in mating than when they did not, and the precopulatory harassment rate was significantly higher on resource territories than on lek territories. After mating on resource territories, harassment dropped to low levels. Thus, resource defenders, but not lek males, seem to employ harassment as a strategy to coerce females to mate against their preference. However, by using various measures of harassment intensity, overall estrous females were found to experience higher harassment levels on lek, and chases by intruders were relatively rare on all territory types. These findings suggest that harassment avoidance is unlikely as an explanation for lek evolution.
Key words: female choice, harassment, lek evolution, male coercion, topi, ungulates.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
When a male and a female in estrus meet, the two individuals are faced with the decision of whether to opt for mating or not. Whether individuals accrue higher fitness benefits by mating than can be anticipated from a continued search for a mating partner depends on the relative quality of the two individuals, as well as on the operational sex ratio in the population (Gibson and Langen, 1996
). As males typically incur lower costs by mating than do females in terms of time and energy spent, a sexual conflict often arises in which males may try to coerce females to mate, and male coercion might thus be considered as the third major force of sexual selection along side male-male competition and female choice (Clutton-Brock and Parker, 1995; Parker, 1970, 1983).
Harassment is probably the most widespread form of coercion in nature (Clutton-Brock and Parker, 1995). In ungulate species in which males employ harassment as a strategy, it has been proposed that female avoidance of harassing males can lead to lek evolution because of relatively low harassment levels on lek (Clutton-Brock et al., 1992, 1993). Off lek, high harassment levels in mixed-sex herds are ascribed to simultaneous harassment from several nonterritorial males, whereas high harassment levels on dispersed territories are attributed to longer chases, overenthusiastic courtship by the territory holder, and higher rates of intrusion by other males. In contrast, it is proposed that estrous females on lek easily escape harassment by switching between the small territories.
The harassment avoidance hypothesis has been addressed in field studies of two lekking ungulate species. Although estrous females in both fallow deer (Dama dama) and lechwe (Kobus leche) were chased more frequently when in mixed-sex herds in nonterritorial areas, no difference was found between chase rates on lek territories and territories off lek (Clutton-Brock et al., 1992; Nefdt, 1995). In lechwe, mounting disruptions were also found to be more common in mixed-sex herds in nonterritorial areas than on lek (Nefdt, 1995), and higher levels of harassment and mounting disruption were found on territories off lek in lekking populations compared with resource territories in nonlekking populations (Nefdt and Thirgood, 1997). However, I have not found any reports of positive results to the critical test of comparing the harassment rates of estrous females on territories on and off lek (Carbone and Taborsky, 1996). To my knowledge, harassment avoidance has never been suggested as a plausible explanation for lek evolution in birds.
The first question addressed in this article is whether harassment is being employed as a strategy in male topi in order to coerce females to mate. If female mating preferences depend on the territory type defended by the male, it could mean that harassment is a relevant strategy only for males on nonpreferred territory types. Hence, I first test if female preference, as measured by mating rate, varies with territory type. For each territory type, I then test if harassment is used as a male strategy, in which case I would expect that (1) female mating probability demonstrates a positive relationship with the harassment levels experienced, and (2) harassment levels subside after mating.
The second question that I address is whether the harassment avoidance model can explain lek evolution in topi. In this case, I would expect that harassment levels experienced by estrous females are lower on leks than elsewhere. As it can be difficult to distinguish whether a chase is harassment (i.e., costly to the female) or whether the female in fact benefits by assessing male display rate, thereby screening a potential mate, I conducted separate analyses on definite harassment events and chases that could be either mate screening or harassment. As part of the analyses, I furthermore investigated the relative contribution of intruding males to harassment levels on territories.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Study area and study animals
Three leks and surrounding resource defense territories (RDTs) were studied in Maasai Mara National Reserve in Kenya between November 1998 and June 2000. Each lek consisted of 11–14 territories. The study area was covered by open rolling grassland with woody vegetation along drainage lines. Wildlife was prolific, and predators included spotted hyena (Crocuta crocuta) and lion (Panthera leo). The topi rut typically lasts 1.5 months and takes place during the long rains between March and May (Bro-Jørgensen, 2001
, 2003). Individual topi were recognized from natural variation, primarily in horn morphology, coloration, and scarring (Gosling and Petrie, 1990). The reliability of individual identification was confirmed by accordance of size measurements from separate sightings (Bro-Jørgensen and Durant, 2003).
Data collection
Fifty-five estrous females, who all mated, were watched focally for a total of 376 h, during which 459 territorial visits took place; however, chases were only recorded for a subset of 44 females. For each focal watch, I selected the first estrous female observed when driving transects through the study site at dawn, typically around 0620 h. The watch continued until 1830 h unless the focal female was lost accidentally or estrus ceased, in which case another estrous female was chosen for focal observation. Thus the mean watch duration became 
min (
). I judged whether or not a female was in estrus based on whether anogenital sniffing caused immediate excitement in males, which was evident from such distinct behaviors as intense displaying, "freezing," tripping at a distance, and spontaneous ejaculation.
During the watch I continuously recorded the territorial location and activity of the female. Interactions with males were recorded by sequence sampling (Altmann, 1974). When a male approached the focal female and she moved away in response, it was recorded as a chase. I divided chases into presumed harassment events (i.e., long chases) and events that could be either screening or harassment (i.e., short chases). I defined a long chase as a chase in which the female either galloped away as a male rushed at her, or persistently moved away in response to an approaching male for at least 6 s. In a short chase, the female moved away from an approaching male for no longer than 5 s; in such cases, it was less clear if a female really attempted to escape.
Chase rates in relation to mating bouts on and off lek
To determine if harassment levels had any influence on whether or not a female decided to mate, I investigated if males had subjected females to higher chase rates when they succeeded in mating than when they did not. I randomly chose 10 central lek males, 10 peripheral lek males, and 10 resource defenders from all the males who had been observed mating with a focal female. For each of these males, I then picked a random mating visit by a focal female and determined the chase rate during the precopulatory period. Subsequently, I determined the chase rate by the same male during a nonmating visit from an estrous female, exactly matching the timing and duration of the interval sampled with the precopulatory period of the mating visit. Correspondingly, in order to determine whether chase rate subsided after mating, I determined the chase rate during the period from mating ceased until the female left the territory (i.e., the postcopulatory period) and the equivalent interval during the nonmating visit to the same male. Randomization was achieved by using a table of random numbers.
Mating and chase rates on and off lek
Statistical comparison of mating rates, chase rates, and chase duration between different locations was done by Friedman tests of related samples to avoid pseudoreplication; multiple comparisons were by Dunnett's tests, testing for significance at the 0.05 level only. With regards to the chase rates and duration, the sample size precluded simultaneous statistical comparison between the three territory types and nonterritorial areas. Therefore, I first compared territory types, including only estrous females who had been watched focally for at least 1 h on all three territory types. Subsequently, I compared lek and nonterritorial areas, including only estrous females who had been observed for a minimum of 20 min in both locations; the reason for reducing the critical time limit was that estrous females only rarely were seen in nonterritorial areas. I conducted the analyses of chase rates and duration for short and long chases separately in order to see if similar patterns emerged whether considering more obvious harassment events (long chases) or possible screening events (short chases).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Female mating preferences
There was a significant difference across territory types in the mating rate of estrous females (Friedman:
females, 
) (Figure 1). Multiple comparisons showed that the mating rate was significantly higher on central lek territories than on both peripheral lek and resource territories, whereas no difference could be demonstrated between the two latter territory types (Dunnett's multiple comparison test: central lek versus peripheral lek territories, 28.08; central lek versus resource territories, 38.22; both, 
; peripheral lek versus resource territories, 10.14, 
). When the mating rate in nonterritorial areas was included in the analysis, the sample size was reduced to nine females, but still there was an overall significant effect of location (Friedman: 
females, 
). Here, multiple comparisons revealed a significant difference between the mating rate on the central lek as compared to nonterritorial areas (Dunnett's multiple comparison test: central lek versus nonterritorial areas, 20.07; ).
|
Is harassment a male strategy?
The chase rate that estrous females experienced on resource territories was significantly higher before mating than during the equivalent time intervals during visits when mating did not occur; in contrast, no significant differences were detected on the two categories of lek territories (Wilcoxon: resource defenders,
; central lek males, 
; peripheral lek males, 
) (Figure 2). The chase rate experienced by estrous females after mating did not differ from the chase rate experienced during matched time intervals of nonmating visits on any of the three territory types (Wilcoxon: resource defenders, 
; central lek males, 
; peripheral lek males, 
; all, 
) (Figure 2).
|
Comparison between the pre- and postcopulatory chase rates for each of the three territory types showed a significant difference in resource defending males only: Here, the rate was significantly higher before mating (Wilcoxon: resource defenders,
; central lek males, 
; peripheral lek males, 
). As a control, I also compared the chase rates experienced during the early and late time intervals during nonmating visits (i.e., the intervals corresponding to pre- and postcopulatory periods during mating visits); here, I found no significant differences (Wilcoxon: resource defenders, 
; central lek males, 
; peripheral lek males, 
; all, ).
Comparison between lek and resource territories showed that the precopulatory chase rate was significantly higher on resource territories; on the other hand, the chase rate during both the early and later phase of nonmating visits was significantly lower on resource territories (Mann-Whitney: 
lek males, 
resource defenders; chase rate during the precopulatory period, 
; during the early period of nonmating visits, 
; during the later period of nonmating visits, 
; during the postmating period, 
).
Testing the harassment avoidance model
The frequency of chases that estrous females experienced varied significantly between territory types for both short and long chases (Friedman: 
estrous females: short chases, 
; long chases, 
) (Figure 3a). Multiple comparisons between territory types showed that frequencies were significantly higher on the central lek (short chases, 17.75/h; long chases, 3.75/h) compared with resource territories (short chases, 7.35/h; long chases, 1.54/h); the only significant difference involving the peripheral lek was that the rate of long chases was significantly lower than on the central lek (peripheral lek: short chases, 10.68/h; long chases, 2.29/h; Dunnett's multiple comparison test, central lek versus RDT: short chases, 16.06; long chases, 11.99; both, ; central versus peripheral lek: long chases, 11.99, ; all others, ).
|
On average, intruding males were responsible for only 0.57% (
) of the chases on the central lek, 5.2% (
) on the peripheral lek and 5.9% (
) on resource territories. There was no significant difference between territory types in either the absolute chase rate of intruders or the chase rate of intruders relative to that of the territory holder (Friedman: rate of short chases, 
; rate of long chases, 
; total chase rate, 
; proportion of all chases by intruder, 
; all, estrous females, ).
Comparison between lek and nonterritorial areas suggested that the chase rate was higher on lek, although the sample size was too small for statistical analysis (central lek: short chases, 27.37/h; long chases, 3.33/h; nonterritorial areas: short chases, 5.13/h; long chases, 0.59/h; 
estrous females) (Figure 3b).
Looking at the time spent in chases, rather than the chase rate, also revealed higher chase intensity on lek, although the differences were significant only as far as short chases were concerned (short chases: central lek, 1.52%; peripheral lek, 0.90%; RDT, 0.64%; long chases: central lek, 0.44%; peripheral, 0.50%; RDT, 0.33%; Friedman: ; short chases, ; long chases, 
) (Figure 4a). Multiple comparisons between territory types showed that the difference between central lek territories and resource territories was significant for short chases (Dunnett's multiple comparison test: short chases, 16.06; ; comparisons involving peripheral lek territories, ).
|
Comparing lek and nonterritorial areas, I found that estrous females spent a larger proportion of their time being chased on lek (short chases: central lek, 2.28%; nonterritorial areas, 0.46%; long chases: central lek, 0.33%; nonterritorial areas, 0.12%;
estrous females) (Figure 4b). |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Judging from mating rate, estrous females preferred to mate with central lek males compared with peripheral lek males who again were preferred to resource defending males. This explains why only resource defenders used harassment as a strategy coercing estrous females to mate: Only resource defenders demonstrated exceptionally high harassment rates before mating, and this cost to females was exacerbated by the longer latency to mating on resource territories (see Bro-Jørgensen, 2002
). Overall, however, estrous females suffered higher harassment levels on lek than off lek, possibly because it was easier to avoid harassing territory holders on large territories, where female herd size typically was larger (Bro-Jørgensen J, unpublished data). This finding renders avoidance of harassment implausible as the cause of lek evolution. The fact that estrous females had a higher mating rate while on the lek also suggests that other forces than harassment avoidance are behind the concentration of mating activity onto leks.
Assessing how costly harassment is to females is not straightforward. First, costly harassment may in some cases be indistinguishable from female screening of male display rate. However, in this study I found higher harassment levels on lek whether I considered more obvious harassment events (long chases) or possible screening events (short chases). Another potential problem is that the cost of harassment to females could be determined by the duration of harassment events rather than by the rate (Carbone and Taborsky, 1996; Clutton-Brock et al., 1996), and it has been suggested that longer chases lead to higher harassment levels off lek (Clutton-Brock et al., 1992; Nefdt, 1995). However, in the present study, I found overall higher harassment levels on lek whether I analyzed harassment rate or duration. No significant difference was found in harassment levels owing to intruders on and off lek. In conclusion, none of the suggestions of higher harassment levels on resource territories (overenthusiastic courtship by the territorial male, longer chases, or higher intrusion rates by other males) seem to apply in the case of topi.
Another ungulate, in which harassment levels can be observed both on and off leks under natural conditions, is lechwe (Nefdt, 1995). As mentioned in the introduction, in lechwe no difference was found between chase rate on lek and on territories off lek, whereas chase rate was significantly higher in nonterritorial areas than on lek. Thus in lechwe, females may leave mixed herds in nonterritorial areas as they come into estrus because of harassment. However, there is little evidence to support that harassment on territories off lek drives females to leks: Although chases were reported to be longer on territories off lek, the difference in the median chase length was minimal (4 versus 5 m). Considering the drastic difference in the ejaculation rate for estrous females on the two territory types (median, 0.33 versus 0 ejaculations/h), the most obvious explanation to my mind is that females are reluctant to accept males defending territories off lek as mating partners to begin with. If estrous females avoid mating with these males solely because of higher harassment levels, it is also hard to see why these males would escalate harassment to such high levels.
In contrast to topi and lechwe, where estrous females can be observed both on and off lek, estrous does in lekking populations of fallow deer are reportedly rarely seen off lek (Clutton-Brock et al., 1992; however, see Thirgood, 1990). Harassment levels off lek have been measured experimentally in fallow deer by herding does off the lek (Clutton-Brock et al., 1992). As in lechwe, higher harassment rates were subsequently found in nonterritorial areas, whereas no difference was detected between lek territories and territories off lek. The harassment in nonterritorial areas was attributed to young males. However, is the reason that these males are avoided as mating partners that they harass more? An experimental study in the species found that females delayed their ovulation by 2 weeks when in the presence of young rather than older males, both categories of males being sexually mature but inexperienced (Komers et al., 1999). As the young males showed insignificant levels of harassment during the delay in breeding, the study shows that harassment avoidance is not the reason for a female preference for older males.
In lekking birds, a study on great snipe (Gallinago media) found that male interference was the most common reason for territory shifts, but there was a tendency for females to be more likely to be chased off territories of preferred males, suggesting that females are choosy in spite of, rather than because of, harassment (Sæther et al., 1999).
I conclude that harassment avoidance is unlikely to be the driving force behind lek evolution in ungulates. In this study I found that although harassment is employed as a strategy by resource defending topi bulls, harassment levels were in fact higher on than off lek. On the other hand, lek evolution by female preference for clustered males finds support in higher female mating rate, as well as female mate competition (Bro-Jørgensen, 2002), on the central lek.
|
ACKNOWLEDGEMENTS |
|---|
I thank S. Durant, M. Gosling, A. Pomiankowski, G. Cowlishaw, C. Roberts, C. Carbone, M. Petrie, W. Sutherland, I. Owens, A. McElligott, and an anonymous referee for their helpful advice. For permissions to do field work, I thank the Office of the President, the Narok County Council, and Kenya Wildlife Service in Kenya; COSTECH and TAWIRI in Tanzania; and the chief park wardens of Maasai Mara National Reserve and Serengeti National Park. Funding was provided by the Danish Research Agency with additional support from the Institute of Zoology, Zoological Society of London, Ottilie Brorson's Travel Fund, and British Airways.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Altmann J, 1974. Observational study of behavior: sampling methods. Behaviour 49:227-265.[Medline]
Bro-Jørgensen J, 2001. Lek-breeding in topi antelopes (Damaliscus lunatus) (PhD dissertation). London: University of London.
Bro-Jørgensen J, 2002. Overt female mate competition and preference for central males in a lekking antelope. Proc Natl Acad Sci USA 99:9290-9293.
Bro-Jørgensen J, 2003. The significance of hotspots to lekking topi antelopes (Damaliscus lunatus). Behav Ecol Sociobiol 53:324-331.[Web of Science]
Bro-Jørgensen J, Durant SM, 2003. Mating strategies of topi bulls: getting in the centre of attention. Anim Behav 65:585-594.[CrossRef]
Carbone C, Taborsky M, 1996. Mate choice or harassment avoidance? A question of female control at the lek. Behav Ecol 7:370-373.
Clutton-Brock TH, Deutsch JC, Nefdt RJC, 1993. The evolution of ungulate leks. Anim Behav 46:1121-1138.[CrossRef]
Clutton-Brock TH, McComb KE, Deutsch JC, 1996. Multiple factors affect the distribution of females in lek-breeding ungulates: a rejoinder. Behav Ecol 7:373-378.
Clutton-Brock TH, Parker GA, 1995. Sexual coercion in animal societies. Anim Behav 49:1345-1365.[CrossRef]
Clutton-Brock TH, Price OF, MacColl ADC, 1992. Mate retention, harassment, and the evolution of ungulate leks. Behav Ecol 3:234-242.
Gibson RM, Langen TA, 1996. How do animals choose their mates? Trends Ecol Evol 11:468-470.[CrossRef]
Gosling LM, Petrie M, 1990. Lekking in topi: a consequence of satellite behavior by small males at hotspots. Anim Behav 40:272-287.[CrossRef]
Komers PE, Birgersson B, Ekvall K, 1999. Timing of estrus in fallow deer is adjusted to the age of available mates. Am Nat 153:431-436.[CrossRef][Web of Science]
Nefdt RJC, 1995. Disruptions of matings, harassment and lek-breeding in Kafue lechwe antelope. Anim Behav 49:419-429.[CrossRef]
Nefdt RJC, Thirgood SJ, 1997. Lekking, resource defense, and harassment in two subspecies of lechwe antelope. Behav Ecol 8:1-9.
Parker GA, 1970. Sperm competition and its evolutionary consequences in insects. Biol Rev 45:525-576.[CrossRef]
Parker GA, 1983. Male quality and mating decisions. In: Mate choice (Bateson P, ed). Cambridge: Cambridge University Press; 141–166.
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.
Thirgood SJ, 1990. Alternative mating strategies and reproductive success in fallow deer. Behaviour 116:1-10.[CrossRef][Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  I. M. Hamilton, M. P. Haesler, and M. Taborsky Predators, reproductive parasites, and the persistence of poor males on leks Behav. Ecol., January 1, 2006; 17(1): 97 - 107. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||