| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Behavioral Ecology Vol. 11 No. 6: 591-596
© 2000 International Society for Behavioral Ecology
Contest behavior in territorial male butterflies: does size matter?
School of Tropical Biology, James Cook University, PO Box 6811, Cairns 4870, Queensland, Australia
Address correspondence to D. Kemp. E-mail: darrell.kemp{at}jcu.edu.au .
Received 20 November 1999; revised 24 February 2000; accepted 10 April 2000.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Although intrasexual contests generally favor bigger or stronger individuals, the relevance of body size to war of attrition-type disputes between weaponless animals such as butterflies is unclear. In this study I aimed to investigate the significance of size in this context by studying territorial contests in Hypolimnas bolina (L.), a species that exhibits consistent seasonal plasticity in body size. In this species adult age is positively correlated with large size in spring but with small size in autumn. This shift allowed independent evaluation of the relevance of each variable (size and age) to intrasexual contest success. Observation of a population of marked individuals indicated that only age appeared important, with the winners of pair-wise contests significantly older than losers in both seasons, and with contests lasting longer when the age difference between the combatants was small. Age was also linked to residency; residents won 99% of all contests. This study suggests that size does not matter in these aerial disputes, but age and residency do. It is not yet possible to determine whether older butterflies are intrinsically better competitors, or whether they simply have greater opportunity to find a vacant territory.
Key words: aggression, war of attrition, nymphalid, Lepidoptera.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Contests between individual males for mating opportunities is an extraordinarily widespread phenomenon, and constitutes an important facet of Darwin's (1871
) theory of
sexual selection. In many instances, males engage in direct physical
struggles, with success generally determined by physical differences between
the combatants (reviewed by Huntingford
and Turner, 1987). Sexual selection in this context is believed to
drive the evolution of larger body size, heavier armor, and larger weapons
(Andersson, 1994;
Darwin, 1871). However, not all
intrasexual conflicts involve direct physical contact. The males of many
butterfly species engage in non-contact aerial contests with rivals for the
ownership of mating territories. These maneuvers, known as "spiral
flights" (Davies, 1978)
or "spinning wheels" (Wickman
and Wiklund, 1983), closely resemble wars of attrition
(Maynard-Smith, 1982). While
body size has been related to the outcome of these contests
(Hernandez and Benson, 1998;
Rosenberg and Enquist, 1991),
the direction of the relationship has been inconsistent, and in other cases,
size has appeared irrelevant to territory ownership
(Knapton, 1985;
Lederhouse, 1982). Because
butterfly contests generally consist of non-contact disputes akin to
stereotyped signaling displays (Davies,
1978; Hernandez and Benson,
1998), it is less obvious why body size (especially large size)
should determine contest outcome, as it does in insect species that grapple or
wrestle for mating opportunities (reviews in
Alcock, 1993;
Alcock and Houston, 1996;
Andersson, 1994). This general
hypothesis is addressed here using the territorial butterfly Hypolimnas
bolina (L.).
Hypolimnas bolina is a tropical and sub-tropical nymphalid species
that is known for its aggressive territorial mate-locating behavior
(Rutowski, 1992). Males of
this species perch in forest clearings and along flight paths as a method of
locating receptive females, and interact with conspecific males in a manner
similar to other butterflies (see Davies,
1978; Hernandez and Benson,
1998; Rosenberg and Enquist,
1991; Wickman and Wiklund,
1983). Intrasexual contests consist either of horizontal chases,
where one male pursues another from the perching area, or escalated contests,
where both males engage in a spiralling maneuver (see
Rutowski, 1992). Casual
observations on an Australian population of this species in the spring 1997
(September) suggested that there may be a size difference between the winners
and losers of male-male contests, in the direction of larger winners. However,
successful males also appeared older (as estimated on the basis of wing
condition) than the butterflies they were able to defeat, which raises the
possibility that size may relate to contest outcome indirectly, due to its
covariance with age.
Body size is related to adult age in this species due to seasonally
predictable phenotypic plasticity in developmental traits that is linked to
reproductive diapause (Kemp and Jones, in
press). Like other adult-diapausing tropical species
(Braby, 1994), individuals that
complete their larval development in late summer/autumn (March-May) eclose as
considerably larger adults than those that develop at other times of the year
(from October-February). Although some adult males from this
"large-size" cohort will compete with existing (older and smaller)
males for the possession of territories throughout the autumn, a proportion
will immediately enter diapause and remain quiescent until early spring
(September). In north Queensland, diapause corresponds to the period of year
in which larval foodplants die off (June-September), and hence, no larval
growth occurs during this time (Kemp DJ, manuscript submitted). When diapause
is terminated, females mate and lay eggs on new season growth
(Kemp, 1998), and these eggs
develop into the first new season cohort of (smaller) males that eclose from
October onward. Since males defend territories for up to 3 months following
diapause (Kemp DJ, unpublished data), large post-diapause males compete with
this new cohort of smaller males throughout late spring/early summer. Hence,
early in the season (October-December), younger males are relatively smaller
than their older adversaries, while late in the breeding season (March-May),
younger males are relatively larger.
Seasonal size plasticity in H. bolina therefore presents a natural experiment, in which the two male attributes, age and body size, vary independently between the seasons. I exploited this phenomenon in this study to determine which of these attributes is consistently related to contest outcome. Since observations in spring 1997 suggested that winners (at that time) were both larger and older than losers, two hypotheses were put forward, each with a straightforward key prediction. The "body size" hypothesis predicted that winners in both spring and autumn should be larger than their opponents, while the "adult age" hypothesis predicted that winners in both seasons should be older. Because the size-age relationship in H. bolina is seasonally inverted, these are mutually exclusive possibilities. The form and duration of intrasexual disputes were also investigated to shed further light on the dynamics of intrasexual disputes in this species.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Study site
This study was conducted along a 770 m transect established adjacent to Freshwater Creek, near Cairns in Queensland, Australia (16°53' S, 145°45' E). The transect was set within a mown track (5-15 m wide) that separated a cultivated sugarcane field from riparian rainforest vegetation. The margin of the track supported several larval foodplants of H. bolina, notably Synedrella nodiflora (L.) (Asteraceae), and the area is a popular perching spot for territorial males.
Sampling methodology
Sampling was conducted on 31 days from 29 March to 22 May, 1998 (the
"autumn round") and on 45 days from 10 October to 29 December,
1998 (the "spring round"), with the same routine followed each
round. For the first 3 days, transects were censused hourly from 0900 h to
1400 h, the period of peak territorial defence
(Rutowski, 1992), where as
many territory residents were caught as possible. Males were marked with a
small number on their ventral hindwing using a green ink marker (Superfine
Stabilo-OH Pen 841, waterproof), assessed for size and age (see below), and
then cooled before release (Kemp and
Zalucki, 1999). Transects were then censused on the days following
the initial marking period to determine the form, duration, outcome, and male
combatants involved in contests at the site. The different levels of
interaction (horizontal chase and escalated contest) are easily
distinguishable (Rutowski,
1992), and the duration of each was timed to the nearest second
using a digital stopwatch. Sampling was limited to sunny, mild conditions (to
control for adverse effects on territorial behavior;
Rutowski, 1992), between 0900
h and 1400 h each day.
Participants of pair-wise contests were identified using binoculars either following or prior to each contest. This task was aided by the tendency of losers to remain within the vicinity of the transect. Notes were also taken on which male was the resident at the territory before the dispute, and the prior behavior of the intruding male. This allowed contests to be classified as either resident-transient (contests between a resident and a male not seen defending a territory prior to the contest) or resident-neighbor (contests between a resident and a male previously seen defending a territory on that day).
In addition, unmarked males seen along the transect were caught, marked, and handled in the manner as previously described. Contests involving males on their day of capture were excluded from the analysis to prevent possible handling effects.
Size and age assessments
Size was assessed by measuring the length from the apex to insertion of the
left forewing to the nearest 0.5 mm
(Hernandez and Benson, 1998).
In order to assess age, the extent of wing wear was subjectively classified on
a five point discrete scale. These wing wear assessments were based on the
degree of fading of darker wing regions (due to systematic scale loss) and
"feathering" of the wingtips, rather than larger chips or areas of
lost wing which may result from one-off incidents. To maintain assessment
accuracy, preserved specimens representing each age class were carried for
comparison with captured individuals.
Discrete age estimates were used in conjunction with date of capture to estimate an approximate date of eclosion for each marked male. This was done by assuming that on average, each wing wear category represented 20 days of life of a male Hypolimnas. This assumption was based on data that suggest males of this species live for a maximum of 3 months in the field (Kemp DJ, unpublished data). Each male was therefore attributed a number of days of adult life on the basis of his wing wear classification, which was subtracted from his date of capture to estimate his eclosion date. This approach gave an approximate age (in days) of each marked combatant in contests observed throughout this study.
Statistical analysis
Following screening of variables for normality (Kolmogorov-Smirnov goodness
of fit; Sokal and Rohlf,
1995), the relationship between size and estimated eclosion date
(age) in each season was investigated using product-moment correlations. This
method was adopted because both variables were subject to error, and least
squares regression techniques are unsuitable. Derived correlation coefficients
were compared by calculating a standard error value for the difference between
the coefficients (ts), then comparing this parameter to
the critical value of t0.05[
] = 1.96
(Sokal and Rohlf, 1995).
Logistic regression was used to model the relationship between contest
outcome and explanatory variables (size and age), with model selection based
on the stepwise elimination procedure outlined by Hardy and Field
(1998). This analysis was
conducted with one male from each pair-wise contest randomly allocated as the
focal male, and contest OUTCOME (the dependent variable) coded as 0 = focal
male lost, 1 = focal male won. The explanatory variables SIZE and AGE were
calculated by subtracting the focal male's age and size from those of his
opponent. The question of "when to escalate?" was also examined
using logistic regression, with escalated contests defined as those containing
a period of spiralling flight (as opposed to those consisting entirely of a
horizontal chase). The SIZE and AGE of the losing male were used as continuous
explanatory variables in this analysis (because the losing male decides if a
contest will escalate), and CONTEXT (0 = resident-transient, 1 =
resident-neighbor) was included as a discrete explanatory variable.
The question of "when to give up?" was addressed using survival
analysis. The duration of the spiraling stage of escalated contests was
analyzed, using a lognormal model, with respect to the SIZE and AGE of the
losing male, the magnitude of the size and age difference between winners and
losers (SIZEDIFF and AGEDIFF), and the CONTEXT under which the contest took
place. Interaction terms between these variables and SEASON were also
included, with the most parsimonious model selected using the stepwise
elimination process (see Hardy and Field,
1998). Survival analysis and logistic regression analyses were
performed using the STATISTICATM computer program.
Although only unique dyads were used in these analyses (only the first of repeat contests between the same two combatants was analyzed), a small proportion (less than 10%) of males were involved in multiple contests. In order to assess the potential bias arising from partially repeated observations, the additional explanatory variable EXPERIENCE was added to the logistic models. In the analysis of contest outcome, this ordinal variable was coded plus 1 for every previously observed win, and minus 1 for every previous loss recorded for each individual male. Similarly, in the analysis of when to escalate, the variable was coded plus 1 for each previous contest that a male was seen to escalate, and minus 1 for every non-escalated contest (horizontal chase). In each analysis, the variable EXPERIENCE tested whether any of the variance in the dependent variable could be explained by the previous experience of males. To account for partially repeated observations in the analysis of contest duration, the giving up times of individual males were averaged across all of the escalated contests in which they were seen to lose. Similarly, in separate comparisons of winner and loser attributes (size and age) in each season, the attributes of winners (of multiple contests) were compared to the mean attributes of the group of males they were observed to defeat.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Seasonal size-age relationships
Totals of 84 autumn and 60 spring males were marked during this study (see Table 1). The size and wing wear of each marked male, as a function of capture date, is indicated by Figure 1. Once both wing wear and date of capture were accounted for (by estimating an eclosion date for each individual), size was related to age as predicted in each season. Forewing size was positively related to estimated eclosion date in autumn (r =.465, n = 84, p <.001), but negatively related to this parameter in spring (r = -.408, n = 58, p <.001), and these correlations differed significantly (ts = 5.36, p <.0001). Older butterflies were therefore larger on average than their younger counterparts in autumn, but this situation was reversed in spring.
|
|
Male-male contests
One hundred thirty-six contests were seen in their entirety (see
Table 1). Of these, 70 were
non-escalated (consisting only of a horizontal chase), while 66 were escalated
and included an initial period of spiraling flight followed by a horizontal
chase in which the loser was pursued from the territory. In line with the
spring 1997 observations (see introduction), winners in spring were on average
larger (paired t14 = 2.79, p <.05) and older
(paired t12 = 4.43, p <.001) than their
corresponding losers. However, winners in autumn were relatively smaller
(paired t28 = -2.34, p <.05) and older (paired
t27 = 2.56, p <.05), and therefore age was the
only seasonally consistent predictor of contest outcome. Winning butterflies
were up to 9.5 mm (28%) larger than their corresponding losers in spring, but
up to 14.0 mm (30%) smaller than losers in autumn. This result supports only
the prediction drawn from the "age-based" hypothesis.
In addition to age, territory residency was a near perfect predictor of contest outcome, with residents winning 86 (99%) of 87 contests in autumn, and all 49 contests in spring. The one clear case of resident loss occurred in autumn, and involved a male (male #16) that had been seen defending the most popular territory (site A) for 6 days. At 0930 h on the morning of his seventh day of residency (26 April, 1998), male #16 was defeated by an apparent newcomer to the transect (male #27) in an escalated contest lasting 137 s. The winner, male #27, was smaller (33.5 compared to 40.5 mm forewing length), and of equal age to male #16 (both estimated 60 days old). After being defeated, the male #16 set up at a nearby territory (20-30 m along the transect), and defended this area over a period of 13 days. During this time he was defeated again by male #27 on one other occasion (interaction lasting 11 s) when he again intruded into male #27's defended area at site A. Later on 26 April (at approximately 1130 h), male #27 was seen mating with a female at site A. This was the only mating observed during the course of this study, and may indicate the value of territorial residence at this site. During the time in which male #27 was in copula, another territorial male, male #23, moved from 100 m along the transect to establish himself at male #27's site. Male #23 was in residence at the site for at least 40 min, and was seen to defend the site from at least one conspecific male intruder (of unknown identity) during this time. Upon his return to the site at approximately 1210 h, male #27 evicted male #23 in an escalated contest lasting 42 s (duration of the spiralling stage = 20 s). Male #27 was later seen to defeat male #23 in three subsequent interactions occurring on 28 and 29 April (contest duration ranging from 9-134 s).
Because prior residency and contest success were virtually the same quantity, logistic regression was used to model the relationship between size, age, and contest outcome/residency. This analysis was performed using contest success as the dependent variable, and the following explanatory variables: SIZE, AGE, SEASON, EXPERIENCE, and two interaction terms: SIZE*SEASON and AGE*SEASON. A model including all variables was significant (G6 = 39.49, n = 69, p <.001), however EXPERIENCE (G1 = 0.63, p >.25), SEASON (G1 = 0.51, p >.25), AGE*SEASON (G1 = 1.37, p >.25), SIZE*SEASON (G1 = 3.39, p >.05) and SIZE (G1 = 3.73, p >.05) were sequentially removed due to their non-significant contribution. The resultant model (overall fit D67 = 32.5, p >.05) included only AGE, with the probability of residency/success increasing as the focal male became older relative to his opponent (G1 = 28.72, p <.001; see Figure 2).
|
When to escalate?
The probability of escalation was examined with respect to the variables
CONTEXT, SIZE, and AGE (of the losing male), SEASON, EXPERIENCE, and the
interaction variables SIZE*SEASON, AGE*SEASON, and
CONTEXT*SEASON. The initial model was significant (G8 =
18.8, n = 85, p <.05), however all variables except
CONTEXT were sequentially removed (G1 < 0.5, p >.25
in all cases). The probability of escalation was therefore seasonally
consistent, and unrelated to either the age, size, or previous experience of
the losing male. The variable CONTEXT accounted for significant deviance
(G1 = 15.0, p <.001) in the final model (overall fit
D83 = 51.4, p >.05), with resident-neighbor contests
more likely to escalate than resident-transient contests (respective
probabilities of escalation - 0.69 [n = 48] and 0.27 [n =
37]).
When to give up?
The most parsimonious lognormal regression model to describe the duration
of spiralling contests included only the variable AGEDIFF, with all others
sequentially eliminated (
12 < 2.9, p
>.05). This model described a significant amount of deviance from that of a
null model (a model including no predictors; 12 =
5.60, n = 38, p <.05). Since AGEDIFF was negatively
correlated to spiraling duration (ß = -.022), longer contests were
observed when the combatants were more equally matched for age.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The clear outcome of this "natural" experiment is that wing size is inconsistently related to the outcome of territorial contests in male H. bolina. This indicates that, unless there is an underlying size-related advantage that undergoes a seasonal inversion, contests in this aggressive species are settled irrespective of asymmetries in body size. This result is biologically significant, because relative combatant size is perhaps the strongest and most consistent proximate determinant of intrasexual contest outcome in animals (reviewed by Andersson, 1994
; Hernandez and Benson,
1998).
Size has previously been correlated with the outcome of territorial
disputes in butterflies, but the direction of this relationship has varied
between species. Rosenberg and Enquist
(1991) found that larger males
of the nymphalid Limenitis weidemeyerii were more successful in
defending and taking over perching territories. Although they also
demonstrated that contest duration was negatively correlated with the
magnitude of the size asymmetry, their results do not discount the hypothesis
that successful males may have also been older. This study on H.
bolina has reiterated that covariance between size and contest outcome
does not necessarily indicate that these variables are causally linked.
Intrinsic correlates of fighting success may themselves covary, and hence,
care must be taken in reaching conclusions regarding the causal merit of
individual attributes.
In a result contrasting with that of Rosenberg and Enquist
(1991), Hernandez and Benson
(1998) reported that
territorial contests among male Heliconius sara (the minor component
of the mating system) were settled in a manner related to small wing size. In
this species, they speculate that large males are less inclined to engage in
territorial disputes due to the risk of injury and potential sacrifice of
residual (future) reproductive value. Size is believed to be important in this
system due to its influence on motivational fighting strategy rather than as a
proximate influence on territorial fighting ability. Hernandez and Benson
(1998) speculate that larger
male H. sara have intrinsically high reproductive value because they
may be competitively superior in gaining perches on about-to-emerge female
pupae (although this may not relate to increased reproductive success; refer
to Deinert et al., 1994). A
similar type of large male advantage has been noted for Jalmenus
evagoras (Elgar and Pierce,
1988), a lycaenid species that also competes for the possession of
eclosing females. However, this rarer form of male-male competition in
butterflies differs markedly from the conspicuous aerial disputes of
territorial species (the stereotyped war of attrition-type contests). The
evidence from the present investigation, taken in concert with other studies
that report either no size-associated territorial advantage
(Knapton, 1985;
Lederhouse, 1982), or
contradictory size-related advantages
(Hernandez and Benson, 1998;
Rosenberg and Enquist, 1991),
suggests that body size has limited proximate bearing on the outcome of
spiraling aerial contests in territorial male butterflies.
The second major outcome of this study is that contest success was related
to both age and prior residency, with residents almost always winning, and
older males winning significantly more contests in both seasons. Prior
residency is often related to contest outcome in butterfly species that
compete for perching territories (see
Davies, 1978;
Rosenberg and Enquist, 1991;
Wickman and Wiklund, 1983).
Residency may be used by male H. bolina as a means to settle the
dispute quickly and in most cases without escalation (an uncorrelated
asymmetry; Maynard-Smith,
1982). Under this scenario, older males may accumulate as
residents (and therefore appear to have a competitive advantage; see
Figure 2) simply because of
their increased opportunity to gain these territories over time. This
possibility is supported by the finding that, as with the territorial
butterfly Papilio polyxenes
(Lederhouse, 1982), contests
between male H. bolina were escalated more often in the
resident-neighbor context, a situation where residency may be confused
(Alcock and Bailey, 1997;
Waage, 1988). However, there
are several potential problems with this interpretation. First, because the
prior intentions of the losers of horizontal chases are not always apparent,
it is possible that many losers in resident-transient chases were not actually
motivated to perch in a territory in the first place (see
Davies, 1978;
Grafen, 1987).
Resident-neighbor contests would be expected to escalate more often than
resident-transient contests simply because a proportion of the latter
interactions may involve males "accidentally" encroaching on an
occupied territory. The second problem is that residency is often correlated
with other asymmetries that may impinge on combatant resource holding
potential (RHP) or motivation to fight
(Austad et al., 1979;
Maynard-Smith, 1982). Body
temperature is one such attribute that may be particularly relevant to
butterfly disputes (Stutt and Willmer,
1998). Residents may win more contests simply because they are
able to maintain their body temperature closer to the optimum, which may allow
greater perseverance in a spiraling maneuver. If body temperature determines
RHP in H. bolina, then escalation should occur most often between
males of similarly high body temperature, which would explain why
resident-neighbor contests were often escalated. However, we would also expect
these contests to last longer than resident-transient contests, which was not
observed. Last, in the few cases where repeat contests were observed
(involving males #16, #23 and #27), the outcome appeared to be consistent
regardless of changes in immediate residency. This result, which has also been
observed by Wickman and Wiklund
(1983) for the nymphalid
Pararge aegeria, and Alcock
(1988) for several territorial
hesperiids, weighs heavily against the uncorrelated asymmetry hypothesis.
Even though age was a less perfect predictor of contest outcome, this
variable may influence the fighting ability of a male butterfly. One possible
explanation is that because the conspicuous spiraling displays are potentially
risky, age may influence motivation. This might apply because age is
negatively related to future reproductive value, and therefore opponents of
different ages would vary in what they have to lose. A younger male, if
injured or killed in a contest, would pay a relatively higher cost than an
older male in terms of lifetime fitness
(Parker, 1974). On this basis,
because they are risking relatively less, older males may be prepared to
spiral for consistently longer, and hence, win more contests against younger
opponents. This strategy is evolutionarily stable
(Grafen, 1987), and
theoretically similar to that advanced by Hernandez and Benson
(1998) for H. sara
with respect to body size (see earlier discussion).
A second possibility is that the ability to survive to old age is determined by the same intrinsic attributes that determine RHP, such as flight ability or agility. Under this scenario, poorer competitors of each successive generation are selected against by "natural" agents such as predation, and hence do not survive to old age. Contests between males should therefore be settled, on average, in favor of the older individual since there is greater probability that he will possess higher RHP. This hypothesis predicts that younger males may win some contests, which was observed in this study, but also that fights will be always won by the faster or more agile male (regardless of age).
If age does determine fighting ability, or motivation to fight, then there
should be a positive relationship between contest duration and the age of the
losing male. This simplistic prediction was not supported: spiraling duration
was negatively related to the magnitude of the age asymmetry (as also found by
Rosenberg and Enquist, 1991
with respect to body size) rather than to age per se. This result raises the
possibility that male H. bolina may assess the age of their opponent
and vary their fighting tactics accordingly
(Enquist and Leimar, 1983).
However, although this strengthens the possibility that age is indeed causally
relevant, the field-based nature of this investigation makes it difficult to
evaluate alternative explanations. Also, precise information on the age of
individual butterflies was not available here, which reduces the sensitivity
of these analyses.
Although the main hypotheses based on age and residency each have a reasonable degree of accordance with the facts, currently exposed evidence does not permit a clear test of all possibilities. As noted in relation to body size, causality cannot be ascribed to individual variables solely on the basis of significant correlation. The value of this "natural" experiment is, therefore, limited to evaluating the relevance of size to war of attrition-type butterfly disputes, and to suggesting primary hypotheses for experimental tests, which are currently underway. Nevertheless, the significance of this outcome should not be understated, because it provides strong, new evidence that size does not substantially influence the outcome of weaponless butterfly disputes.
|
ACKNOWLEDGEMENTS |
|---|
I am indebted to Scott Field, Ian Hardy, and Rhondda Jones for providing advice on critical aspects of this research. I also thank Woodruff Benson, Rob Brooks, Brian McArdle, Simon Robson, Richard Rowe, Ronald Rutowski, and Christer Wiklund. This research was supported by an Australian Postgraduate Research Award.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Alcock J, 1988. The mating system of three territorial butterflies in Costa Rica. J Res Lepid 26: 89-97.
Alcock J, 1993. The effects of male body size on territorial and mating success in the landmark-defending fly Hermetia comstocki (Stratiomyidae). Ecol Entom 18: 1-6.
Alcock J, Bailey WJ, 1997. Success in territorial defence by male tarantula hawk wasps Hemipepsis ustulata: the role of residency. Ecol Entom 22: 377-383.
Alcock J, Houston TF, 1996. Mating systems and male size in Australian Hylaeine Bees (Hymenoptera: Colletidae). Ethology 102: 591-610.
Andersson M, 1994. Sexual selection. Princeton, New Jersey: Princeton University Press.
Austad SN, Jones WT, Waser PM, 1979. Territorial defence in speckled wood butterflies: why does the resident always win? Anim Behav 27: 960-961.
Braby MF, 1994. Phenotypic variation in adult Mycalesis Hubner (Lepidoptera: Nymphalidae: Satyrinae) from the Australian wet-dry tropics. J Aust Entomol Soc 33: 327-336.
Darwin C, 1871. The descent of man and selection in relation to sex. London: Murray.
Davies NB, 1978. Territorial defence in the speckled wood butterfly (Pararge aegeria): the resident always wins. Anim Behav 26: 138-147.
Deinert EI, Longino JT, Gilbert LE, 1994. Mate competition in butterflies. Nature 370: 23-24.
Elgar MA, Pierce NE, 1988. Mating success and fecundity in an anttended Lycaenid butterfly. In: Reproductive success: individual variation in complex breeding systems (Clutton-Brock TH, ed). Chicago: University of Chicago Press; 59-75.
Enquist M, Leimar O, 1983. The evolution of fighting behaviour: decision rules and assessment of relative strength. J Theor Biol 02: 387-410.
Grafen A, 1987. The logic of divisively asymmetric contests: respect for ownership and the desperado effect. Anim Behav 35: 462-467.
Hardy ICW, Field SA, 1998. Logistic analysis of animal contests. Anim Behav 56: 787-792.[Web of Science][Medline]
Hernandez MIM, Benson WW, 1998. Small-male advantage in the territorial tropical butterfly Heliconius sara (Nymphalidae): a paradoxical strategy? Anim Behav 56: 533-540.[Web of Science][Medline]
Huntingford F, Turner A, 1987. Animal conflict. New York: Chapman & Hall.
Kemp DJ, 1998. Oviposition behaviour of post-diapause Hypolimnas bolina (L.) (Lepidoptera: Nymphalidae) in tropical Australia. Aust J Zool 46: 451-459.
Kemp DJ, Jones RE, in press. Phenotypic plasticity in field populations of the tropical butterfly Hypolimnas bolina (L.) (Nymphalidae). Biol J Linn Soc.
Kemp DJ, Zalucki MP, 1999. Method of handling affects postcapture encounter probabilities in Hypolimnas bolina (L.) (Lepidoptera: Nymphalidae). J Lepid Soc 53: 138-141.
Knapton RW, 1985. Lek structure and territoriality in the chryxus arctic butterfly, Oeneis chryxus (Satyridae). Behav Ecol Sociobiol 17: 389-395.
Lederhouse RC, 1982. Territorial defense and lek behaviour of the black swallowtail butterfly, Papilio polyxenes. Behav Ecol Sociobiol 10: 109-118.
Maynard-Smith J, 1982. Evolution and the theory of games. Cambridge: Cambridge University Press.
Parker GA, 1974. Assessment strategy and the evolution of fighting behaviour. J Theor Biol 47: 223-243.[Web of Science][Medline]
Rosenberg RH, Enquist M, 1991. Contest behaviour in Weidemeyer's admiral butterfly Limenitis weidemeyerii (Nymphalidae): the effect of size and residency. Anim Behav 42: 805-811.
Rutowski RL, 1992. Male mate-locating behaviour in the common eggfly, Hypolimnas bolina (Nymphalidae). J Lepid Soc 46: 24-38.
Sokal RR, Rohlf FJ, 1995. Biometry, 3rd ed. New York: W.H. Freeman & Co.
Stutt AD, Willmer P, 1998. Territorial defence in speckled wood butterflies: do the hottest males always win? Anim Behav 55: 1341-1347.
Waage JK, 1988. Confusion over residency and the escalation of damselfly territorial disputes. Anim Behav 36: 586-595.
Wickman P.-O, Wiklund C, 1983. Territorial defence and its seasonal decline in the speckled wood butterfly (Pararge aegeria). Anim Behav 31: 1206-1216.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. J. Fuxjager and C. A. Marler How and why the winner effect forms: influences of contest environment and species differences Behav. Ecol., November 13, 2009; (2009) arp148v1. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. R. Foote, L. P. Fitzsimmons, D. J. Mennill, and L. M. Ratcliffe Male chickadees match neighbors interactively at dawn: support for the social dynamics hypothesis Behav. Ecol., November 1, 2008; 19(6): 1192 - 1199. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Fischer, J. Perlick, and T. Galetz Residual reproductive value and male mating success: older males do better Proc R Soc B, July 7, 2008; 275(1642): 1517 - 1524. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Bergman, K. Gotthard, D. Berger, M. Olofsson, D. J Kemp, and C. Wiklund Mating success of resident versus non-resident males in a territorial butterfly Proc R Soc B, July 7, 2007; 274(1618): 1659 - 1665. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. J. Kemp Butterfly contests and flight physiology: why do older males fight harder? Behav. Ecol., July 1, 2002; 13(4): 456 - 461. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||