TERRITORY TENURE IN A LIBELLULID DRAGONFLY
Animals fight
over scarce resources. In each fight, three asymmetries may determine the
winner of a fight: value of the contested resource, fighting ability (resource
holding potential, RHP), or an asymmetry uncorrelated with physical characters,
such as ownership (e.g.: Maynard Smith & Parker 1976; Hammerstein 1981,
Hammerstein & Parker 1982). The relative importance of asymmetries
in resource holding potential (RHP) and resource value (RV) on single contest
outcomes has been studied in many taxa (Austad 1983; Verrell 1986; Beletsky
and Orians 1987, 1989; Bjorklund 1989; Yokel 1989; Dugatkin and Ohlsen
1990; Dugatkin and Biederman 1991). All of these studies assumed or showed
that RHP is related to size or weight of the contestant. Resource value
was variously defined as: presence of a female during the fight (Yokel
1989), pairing status (Bjorklund 1989), length of time the territory had
been held (Beletsky and Orians 1987, 1989), female size (Austad 1983; Verrell
1986), availability of territories (Riechert 1979), or expected feeding
regime (Dugatkin and Ohlsen 1990; Dugatkin and Biederman 1991). In some
cases resource value was more important than RHP in determining who won
a fight (Krebs 1982; Beletsky and Orians 1987, 1989; Bjorklund 1989; Yokel
1989) whereas in other cases, resource value affected contest duration,
but not contest outcome (Riechert 1979; Austad 1983; Verrell 1986; Dugatkin
and Ohlsen 1990; Dugatkin and Biederman 1991).
In contrast
to the extensive research regarding single contests, questions about repeated
contests are essentially unexplored. However, a territorial individual
must win repeated contests against a number of different intruders in order
to maintain its territory. A male may eventually lose his territory when
the value he places on the resource changes, or when his resource holding
potential changes. Resource value can change as the environment changes,
or the territory owner's evaluation of resource value can change independent
of external changes (Krebs 1982). Resource holding potential can change
if it is related to age or energy levels. For example, energy intake and
energy use may affect the presence of males at mating sites in elephant
seals (Mirounga angustirostris, Le Boeuf & Reiter 1988) and
in sage grouse (Centrocercus urophasianus, Vehrencamp et al. 1989).
Some territorial
dragonflies (Odonata: Anisoptera) maintain territories for only a few hours
each day (Fitzpatrick and Wellington 1983). When more males are present,
the amount of time a male dragonfly spends fighting increases, resulting
in shorter territory tenure (Kaiser 1974; Koenig 1990). Length of territory
tenure can also be affected by resource value, defined as rate of female
arrival (Koenig 1990). Male Pachydiplax longipennis appear to ingest
approximately the same amount of energy each day that they expend on territories,
suggesting that energy might be an important factor in their territoriality
(Fried & May 1983).
If males lose
fights because they run out of energy (Marden & Waage 1990), a male
that fights more should run out of energy faster and lose his territory
sooner than a less active male, all else being equal. Similarly, a male
with more energy reserves may take longer to run out of energy. If resource
value determines the outcome of contests, then a male that is able to copulate
frequently should continue to invest in winning fights and remain on his
territory longer than a male that encounters and copulates with few females,
again, all else being equal.
In nature
it would be difficult to vary one factor (energy expenditure, energy intake
or female arrival rate) while holding other factors equal. Therefore I
manipulated feeding history, male density, and territory value (rate of
female arrival) for P. longipennis in large flight cages to test
between two hypotheses: (1) energy balance determines how long a male can
remain territorial and (2) female arrival and mating success determine
how long a male stays territorial.
METHODS
Pachydiplax
longipennis is a medium-sized (total length 28-45 mm) territorial dragonfly
abundant at ponds and streams in the eastern United States. Gravid females
come to the water to oviposit and usually to mate. They are approached
by territorial males at the water with whom they copulate for about half
a minute. After copulation the male releases the female and hovers near
her while she oviposits. Oviposition takes only one or two minutes, after
which the female leaves the pond and the male returns to defending his
territory. Fights involve one male, abdomen raised, chasing another. Occasionally
fights escalate, and the two males fly rapidly upward at a 40 degree angle,
to approximately 10 meters into the air. Males that leave their territories
during the day almost always leave after a fight. Males leave their territories
at night to roost in the treetops. Both males and females forage away from
the pond.
I carried
out the following research between 15 April and 24 August 1991 and between
15 May and 31 July 1992 at the University of Florida Research and Education
Center (UFREC) in Welaka, Putnam County, FL. At the UFREC I built two large
(10 x 20 x 3 m) enclosures (Dunham 1994). One cage, which contained part
of a larger pond, held males prior to experiments. The experimental cage
contained a man-made pond (5 x 10 m). This pond was marked at 1 m intervals.
The experimental cage was split with a mesh wall into two halves of equal
size, each containing half of the pond. Pachydiplax behave normally
in such a cage, although they survive significantly longer than free-living
individuals (pers. obs.). Sympetrum also behaved normally in similar
enclosures (Michiels 1989).
Preliminary
observations indicated that on the unenclosed pond no more than two male
Pachydiplax were territorial at a time. In contrast, the same pond
supported as many as four territorial males at the same time during the
experiments.
Each adult
male was marked, weighed to the nearest .1 mg, and the left hindwing was
measured to the nearest 0.1 mm before introduction to the cage. Females
were also individually marked.
Energy reserves
can determine contest outcome in some odonates (Marden and Waage 1990).
Therefore, I sacrificed all males at the end of each experiment to determine
two measures of energy reserves: fat and gut contents. Fat content should
be a good estimate of long-term energy balance. Gut content should be a
good estimate of energy intake in the previous 24 hours (Fried and May
1983). I immediately removed and weighed the gut, and returned it to the
abdominal cavity. I then weighed each body part separately (head, thorax,
abdomen, wings and legs). Fat was extracted from dried body parts for 8
h with chloroform in a Soxhlet apparatus (Marden 1989).
Weather conditions
affect activity in these ectotherms (e.g.: Heinrich and Casey 1977; Pezalla
1979; Hilton 1983). I therefore recorded light level (in foot candles),
cloud cover, and temperature every 15 min throughout the day.
I recorded
the identity and position of all males at the enclosed pond every 15 minutes
between 0800 and 1800 in order to estimate tenure in hours. Between 20
May and 16 August 1991 I recorded tenure for 185 males on 37 days for a
total of 277 male-days of tenure. Between 13 June and 19 July 1992 I recorded
tenure for 87 males on 18 days for a total of 167 male-days of tenure.
Weather conditions and treatment changed from day to day. Therefore, tenure
for an individual male on a given day was considered to be independent
from that individual's tenure on other days (Koenig 1990).
I caught a
272 males as adults. For each experiment, males were allowed to settle
on territories on both sides of the enclosure. Experiments started the
day after capture. Two different treatments were presented on a given day,
one in each half of the cage. Males were used in only one experiment each,
but experienced more than one treatment during an experiment.
Experiment #1 - Effects of Density and Female Presence
on Tenure
I used a repeated-measures
design with environmental variables as covariates. Treatments were presented
in random order, each lasting one day. The experiment lasted two days.
To reveal the effects of density and female presence on territory tenure,
I recorded tenure (in hrs) of males exposed to varying male densities in
the presence or absence of females. Male density was expected to affect
energy expenditures by increasing activity (Koenig 1990), and female presence
was expected to increase territory value by increasing encounter rate and
mating frequency.
I manipulated
territory value by releasing as many gravid females as possible (up to
11) into one side of the experimental cage and none into the other. The
ratio of males to females ovipositing ranged from 1:1 to 11:1. In most
cases, males experienced either female presence or female absence throughout
both days of a two-day experiment. In one instance females were added to
the cage on the second day of the experiment. During the four months of
study, each side of the cage contained females an equal number of days.
Not all females present in the cage oviposited on a given day. On days
when females were present, an average of 1.35 + 1.38 different females
oviposited in the cage (range = 0 to 6). That is, 30.1 + 27.0% of
the females in the cage oviposited. Females always mated before ovipositing,
and often visited all territories in the cage before mating.
The density
of males in the cage ranged between 1 and 13, in order to manipulate activity
level. To test whether the density manipulation affected activity level,
I watched focal males for 15 min each from 27 June to 16 August. I discontinued
observations if the focal male tried to escape from the cage or if I lost
sight of him in the undergrowth. Observations ranged from 3 to 15 min.
I conducted focals on all males at the pond in turn. I timed the duration
of flight, patrols, chases, and copulations during each focal watch. I
also recorded the identity of females copulating during focals. In all,
I conducted 234 focals on 65 males at densities ranging from 1 to 13 males
per cage half. I also recorded 15 focal watches on 6 males in the wild.
Males experienced
a different density on each day of the two-day experiment, in random order.
Density of males was decreased by removing randomly chosen males at 1900
hr on the first day, or increased by adding males before 0900 hr on the
second day of the experiment.
Experiment #2 - Effects of Repeated Feeding, Male Density,
and Female Presence
In 1992 I
manipulated feeding history of males, as well as manipulating male density
and female presence. Male Pachydiplax were hand fed by gently introducing
a teneral Enallagma sp. of known weight to the mouthparts. Males
were then allowed to perch on a stick in a bucket. Any remains of the meal
were collected and weighed. All fed males consumed at least some food.
Control males were handled similarly. The Enallagma weighed 17.6
+ 2.24 mg (N=217), of which fed males consumed 7.9 + 8.65 mg.
Males normally eat 7.86-9.95 mg in an entire day (Fried and May 1983).
To maximize the effect of feeding, experimental males were fed for three
days in a row prior to the experiment, as well as daily during the four-day
experiment. All males were allowed to feed freely in the enclosure.
Female presence
was manipulated as described for experiment #1. As many as 13 females were
added to a cage half. The operational sex ratio (males territorial to females
ovipositing that day) ranged from 4:1 to 8:9.
Male density
was adjusted the evening before each day of the experiment. Males that
disappeared overnight were replaced before 0800 h the day of the experiment.
Four males were used in each low density treatment, and 8 males were used
in each high density treatment. Once again I used a repeated-measures design.
Treatments were presented in random order, such that each male experienced
4 of 8 combinations of males density, female presence, and feeding treatment.
The 45 males used in this experiment contributed a total of 95 male-days.
RESULTS
The cage environment and my manipulations significantly
affected fighting behaviour (Table I). In the cage, density was positively
but weakly related to the time focal territorial males spent fighting (r2
= 0.044, N = 185, p = 0.004) and to the number of fights during focals
on territorial males (r2 = 0.090, N = 185, p = 0.0001).
Total time active and density were not related (r2
= 0.001, N = 221, p = 0.71).
In the cage,
males fought an average of 0.48 + 1.29 times during 15 minute focals
(N=221, range = 0 to 12 times ) for an average of 4.3 + 15.7 sec
fighting (range = 0 to 190 sec). At high density in the wild, males fought
an average of 5.0 + 3.3 times during 15 minute focals (N=55, range
= 0 to 12) for an average of 58 + 69 sec fighting (range = 0 to
394 sec). Males fought significantly less in the cage than in the wild
(t=5.738, p < 0.0001).
In 1991, the
average tenure in one day for all males in the cage was 2.47 + 2.48
h (range = 0 to 10 h). In 1992, average tenure for caged males was 2.98
+ 2.43 h (range = 0 to 8.25 h). Tenure was not significantly different
in 1991 from tenure in 1992 (t = 1.74, p > 0.05). Despite the higher level
of activity in the wild, tenure in the wild (2.86 + 2.25 h) did
not differ from in the cage (t=0.77, df = 275, p > 0.05). Males with tenure
of zero were considered non-territorial. An average of 83.1 + 21.09%
of males present were territorial on any given day.
Experiment #1 - Effects of Density and Female Presence
on Tenure
Only density
had a significant effect on mean tenure (Table II). Recorded weather conditions
had no effect on territory tenure. Although female presence did not affect
tenure, females did affect territorial males. The effect of density on
time spent fighting, though small, was stronger in the presence of females
(slope of regression = 0.09; 95% confidence limit = 0.04 - 0.14) than in
their absence (slope 0.017; 95% confidence limit = 0.001 - 0.034). Similarly,
the effect of density on the number of fights was stronger when females
were present (slope = 0.141; 95% confidence limit = 0.081 - 0.201) than
in their absence (slope = 0.033; 95% confidence limit = 0.00 - 0.070).
In addition, males in the presence of females had less fat/total body mass
at the end of the day (1.5% + 1.0, N = 52) than males with no females
present (2.3% + 1.8, N = 50; p=0.0065).
Experiment #2 - Effects of Feeding, Male Density, and
Female Presence
Feeding increased
energy reserves in fed males. Males that were fed were fatter (N=9, mean
= 5.6 mg total fat content + 1.92) at the end of the experiment
than males that were not fed (N=10, mean = 3.1 + 1.29, t=3.4, p=0.004).
Female presence
had no effect, but feeding and male density did affect tenure duration
(Table III). Males that had been hand fed stayed on their territories significantly
longer than control males (fed males: 3.7 + 2.51 h, control males:
2.3 + 2.17 h, t=2.86, p = 0.005). Fed males began territory defense
earlier than unfed males on average, though this difference was not quite
significant (fed: 11.1 h + 1.2, unfed: 11.8 + 1.9, t=1.953,
p=0.054). As additional evidence that feeding affected tenure, length of
tenure significantly increased with number of days a male was fed, independent
of male density or female presence (r2=0.065, p=0.0121,
F=6.59). At high male density, individual males maintained territories
for a shorter time than at low male density (Table III). Interactions were
not significant.
DISCUSSION
Despite extreme
variation in the system and differences between conditions in the enclosure
and conditions in the wild, energy levels and male density significantly
affected territory tenure in male P. longipennis.
The length
of time males can defend territories depends on their resource holding
potential (here, energy balance at the beginning of the day). Territorial
behaviour carries a double cost - males do not feed while territorial,
and they expend energy at a high rate. Fed males were able to maintain
territory defense longer than unfed males. Fried and May (1983) originally
suggested that P. longipennis may be energy limited based on their
estimates of energy intake and energy consumption. This species will forage
faster when more prey are available (Baird, unpubl. data), further indicating
energy limitation. The current study experimentally shows that territorial
males are indeed energy limited. Alternatively, males are time limited,
in the sense that they cannot simultaneously forage (take in energy) and
behave territorially (expend energy).
Male density
acted to decrease tenure duration by increasing the number and length of
fights during the day, thereby presumably increasing energy use. Total
activity did not significantly increase as density increased, suggesting
that fights take time away from other activities. Territory duration is
also negatively correlated with male density and with time spent chasing
in the libellulid dragonfly Plathemis lydia (Koenig 1990).
Resource value,
here defined as female presence, did not affect territory tenure. However,
males are more sensitive to the presence of other males and use up their
fat reserves when females are present. In the cage, female arrival rate
was very low (mean 1.35/day/30 m shoreline), and time of arrival was unpredictable,
so that territory tenure may not have been a strong predictor of reproductive
success. However, even with high female arrival rate (8.07/day/30 m), in
P. lydia there is no correlation between territory tenure and mating
success (Table 2, Koenig 1990).
In mating
systems defined by limited access to areas where females gather and mate,
males are not able to spend all of their time in pursuit of females. The
observed ability of male dragonflies to display aggression longer when
they consume more energy fits a wider taxonomic trend. Male barking treefrogs
(Hyla gratiosa) who were artificially fed before arrival at a pond
stayed in the chorus longer, thereby acquiring more mates (Murphy 1994).
Male sage grouse (Centrocercus urophasianus) expend up to 17.4 times
their basal metabolic rate while displaying on a lek (Vehrencamp, et al.
1989). Males who are very active on the lek appear to forage further from
the lek than less active males, suggesting the importance of foraging to
continued active display. Where the resource defended is spatially limited
access to gravid females, resource holding potential can be defined in
terms of energetic constraints in a wide variety of taxa, both cold-blooded
and warm-blooded, both vertebrate and invertebrate.
Acknowledgements
- I would like to thank J. Waage for his patient assistance. My ideas have
benefited greatly from discussions with him and with L. Brooks, S. Gaines,
and D. Morse. M. May, J. Marden, O. Fincke, R. Bohr, L. Wolf, and B. Anholt
made helpful suggestions on the manuscript. Support for this work was provided
by Sigma Xi and the Animal Behaviour Society, as well as the Graduate School
of Brown University. Preparation of this MS was supported by NSF grant
#DEB 9107078. The University of Florida at Gainesville allowed me to use
their facilities (ACF & UFREC), for which I thank them. Special thanks
to M. Miller, O. Reese, D. Schultz, M. Stewart, and A. Taylor for all their
help at the former UFREC.
LITERATURE CITED
Austad S.N. 1983. A game theoretical interpretation of male combat in the bowl and doily spider (Frontinella pyramitela). Anim. Behav. 31: 59-73.
Beletsky L.D. & G.H. Orians. 1987. Territoriality among male red-winged blackbirds. II. Removal experiments and site dominance. Behav. Ecol. Sociobiol. 20: 339-349.
Beletsky L.D. & G.H. Orians. 1989. Territoriality among male red-winged blackbirds. III. Testing hypotheses of territorial dominance. Behav. Ecol. Sociobiol. 24: 333-339.
Björklund M. 1989. Male contests in the scarlet rosefinch (Carpodacus erythrinus) in relation to asymmetries in resource holding power and mating status. Behav. Ecol. Sociobiol. 25: 137-140.
Dugatkin L.A. & L. Biederman. 1991. Balancing asymmetries in resource holding power and resource value in the pumpkinseed sunfish. Anim. Behav. 42: 691-692.
Dugatkin L.A. & S.R. Ohlsen. 1990. Contrasting asymmetries in value expectation and resource holding power: Effects on attack behaviour and dominance in the pumpkinseed sunfish, Lepomis gibbosus. Anim. Behav. 39: 802-804.
Dunham M.L. 1994. An easily constructed temporary cage for studying small animals in the field. Fla. Entomol. 77: 505-508.
Fitzpatrick S.M. & Wellington W.G. 1983. Insect territoriality. Can. J Zool. 61: 471-486.
Fried C.S. & May M.L. 1983. Energy expenditure and food intake of territorial male Pachydiplax longipennis (Odonata: Libellulidae). Ecol. Entomol. 8: 283-292.
Hammerstein P. 1981. The role of asymmetries in animal contests. An. Beh. 29: 193-205.
Hammerstein P. & Parker G. A. 1982. The asymmetric war of attrition. J. theor. Biol. 96: 647-682.
Heinrich B. & Casey T.M. 1978. Heat transfer in dragonflies: 'fliers' and 'perchers'. J. exp. Biol. 74: 17-36.
Hilton D.F.J. 1983. Territoriality in Libellula julia Uhler (Anisoptera: Libellulidae). Odonatol. 12: 115-124.
Kaiser H-J. 1974. Verhaltensgefüge und Temporialverhalten der Libelle Aeschna cyanea (Odonata). Z. Tierpsychol. 34: 398-429.
Koenig W. D. 1990. Territory size and duration in the white-tailed skimmer Plathemis lydia (Odonata: Libellulidae). J Anim. Ecol. 59: 317-333.
Krebs J. R. 1982. Territorial defence in the great tit (Parus major): do residents always win? Behav. Ecol. Sociobiol. 11: 185-194.
Le Bouef B. J. & Reiter J. 1988. Lifetime reproductive success in northern elephant seals. In: Reproductive Success: Studies of Individual Variation in Contrasting Breeding Systems. (Clutton-Brock T.H., ed.). Chicago: University of Chicago Press.
Marden J.H. 1989. Bodybuilding dragonflies: Costs and benefits of maximizing flight muscle. Physiol. Zool. 62: 505-521.
Marden J.H. & Waage J.K. 1990. Escalated damselfly territorial contests are energetic wars of attrition. Anim. Behav. 39: 954-959.
Maynard Smith J. 1974. The theory of games and the evolution of animal conflicts. J. theor. Biol. 47: 209-222.
Maynard Smith J. & Parker G. A. 1976. The logic of asymmetric contests. Anim. Behav. 24: 159-175.
Michiels N. 1989. Populatie- en gedragsecologie van de Zwarte Heidelibel Sympetrum danae (Sulzer) (Odonata: Libellulidae). (Ph.D. dissertation). Antwerp, Belgium: University of Antwerp.
Moore A.J. 1987. The behavioral ecology of Libellula luctuosa (Burmeister) (Anisoptera: Libellulidae). I. Temporal changes in the population density and the effects on male territorial behavior. Ethology 75: 246-254.
Murphy C. G. 1994. Determinants of chorus tenure in barking treefrogs (Hyla gratiosa). Behav. Ecol. Sociobiol. 34: 285-294.
Parker G.A. 1974. Assessment strategy and the evolution of fighting behavior. J. theor. Biol. 47: 223-243.
Parker G.A. & Rubenstein D. I. 1981. Role assessment, reserve strategy, and acquisition of information in asymmetric animal conflicts. Anim. Behav. 29: 221-240.
Pezalla V.M. 1979. Behavioral ecology of the dragonfly Libellula pulchella Drury (Odonata: Anisoptera). Am. Midl. Nat. 102: 1-22.
Riechert S. E. 1979. Games spiders play. II. Resource assessment strategies. Behav. Ecol. Sociobiol. 6: 121-128.
Shelly T. E. 1982. Comparative foraging behavior of light- versus shade-seeking adult damselflies in a lowland neotropical forest (Odonata: Zygoptera). Physiol. Zool. 55: 335-343.
Sherman K. J. 1983. The evolution of reproductive strategies in a dragonfly, Pachydiplax longipennis. (Ph.D. dissertation). Ithaca, NY: Cornell University.
Southwood 1968. Ecological Methods with Particular Reference to the Study of Insect Populations. Methuen: London.
Vehrencamp S.L., Bradbury J.W. & Gibson R. M. 1989. The energetic cost of display in male sage grouse. Anim. Behav. 38: 885-896.
Verrell P. A. 1986. Wrestling in the red-spotted newt (Notophthalmus viridescens): Resource value and contestant asymmetry determine contest duration and outcome. Anim. Behav. 34: 398-402.
Yokel D.A. 1989. Payoff asymmetries in contests among
male brown-headed cowbirds. Behav. Ecol. Sociobiol. 24: 209-216.
Table I. Means and standard deviations of A. number of fights/15 min focal (N) and B. number of seconds fighting/15 min focal (N).
A. Number of fights/15 min focal period.
| low
(1-3 males) |
high
(7+ males) |
natural
conditions |
|
| _________________________________________________________________________________ | |||
|---|---|---|---|
| Females present | 0.12+0.392 | 1.15+2.040 | 5.0+3.3 |
| (90) | (66) | (55) | |
| Females absent | 0.21+0.787 | 0.43+0.959 | |
| (28) | (37) | ||
B. Number of seconds fighting/focal period.
| low
(1-3 males) |
high
(7+ males) |
natural
conditions |
|
| _________________________________________________________________________________ | |||
|---|---|---|---|
| Females present | 1.2+4.4 | 10.9+26.7 | 58+69 |
| (90) | (66) | (55) | |
| Females absent | 0.1+0.3 | 3.3+7.6 | |
| (28) | (37) | ||
Table II. Effects of female presence and male density
(number of active males in cage) and their interaction on length of territory
tenure (in hrs). Environmental variables are covariates. Males that were
not territorial are not included in the evaluation of tenure. Each sample
is one male-day. R2= 0.061.
Treatment
df
Mean Square P-Value
________________________________________________
Female presence
1
10.580
0.1637
Male density
1
35.612
0.0112
Female presence
1
12.631
0.1283
by Male density
Covariates:
Average cloud
1
3.618
0.4145
Average light
1
4.045
0.3883
Average temperature
1
1.075
0.6563
Residual
165 5.405
Female presence
N
Mean Std. Dev.
_______________________________________
Females present
87
3.28
2.08
No females
82
3.40
2.63
Table III. Female presence, male density (number of active
males in cage), feeding history (fed every day), and their interactions
versus length of territory tenure (in hrs). Each sample is one male-day.
R2= 0.150. Means table presented underneath.
____________________________________________________
Source of variation
df Mean square
F
P
____________________________________________________
Feeding history (A)
1 50.894
9.379 0.0029
Female presence (B)
1 3.376
0.622 0.4324
Male density (C)
1 33.168
6.112 0.0154
AB
1 0.006
0.001 0.9727
AC
1 2.772
0.511 0.4767
BC
1 0.307
0.056 0.8127
ABC
1 0.092
0.017 0.8969
Residual
87 5.427
____________________________________________________
Treatment
N
Mean
Std. Dev.
________________________________________
Fed
47
3.676
2.505
Not fed
48
2.302
2.173
Females present
49
3.107
2.421
Females absent
46
2.848
2.462
Low density
32
3.766
2.584
High density
63
2.583
2.268