Behavioral Ecology Advance Access originally published online on August 16, 2007
Behavioral Ecology 2007 18(6):1021-1028; doi:10.1093/beheco/arm071
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Environmental heterogeneity and alternative mating tactics in the damselfly Protoneura amatoria
Department of Ecology and Evolutionary Biology, University of California, Los Angeles, 621 Charles E. Young Drive South, Los Angeles, CA 90095, USA
Address correspondence to B. Larison. E-mail: blarison{at}ucla.edu.
Received 3 February 2007; revised 12 July 2007; accepted 12 July 2007.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONFUNDINGREFERENCES |
|---|
Conditional male mating strategies have been studied extensively in relation to male attributes, such as size and resource-holding potential, but few studies have considered the effects of environmental heterogeneity on the use of alternative mating tactics. In some systems, environmental heterogeneity may be the key to understanding the evolution and maintenance of such polyphenisms. I examined the influence of the physical environment on the use of alternative tactics by the damselfly Protoneura amatoria. Male P. amatoria reversibly use 2 tactics to gain matings: 1) sit and wait in the canopy for passing females or 2) hover over the water and attempt to grab females that are ovipositing in floating debris. Observations in 3 streams indicated that the use of the hovering tactic was greater under high-light than low-light conditions and at higher densities of ovipositing females. The density of ovipositing females was correlated with both the light conditions and the availability of oviposition substrate, indicating that physical factors exert indirect as well as direct influences on tactic use. Experimental manipulations showed that both males and females responded directly to light conditions and suggested that males responded directly to the density of ovipositing females. These results can be explained largely in terms of the cues and constraints inherent in different light environments. Thus, the conditional mating strategy of P. amatoria appears to have evolved in response to, and been maintained by, fine-scale variation in the physical environment. These findings are discussed in relation to flight dynamics and predation risk.
Key words: Odonata, phenotypic plasticity, Protoneuridae, Zygoptera.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONFUNDINGREFERENCES |
|---|
The influence of environmental heterogeneity on alternative mating tactics has gone largely unexamined in spite of recent theoretical models that suggest a role for it in the evolution of conditional mating tactics and other plastic traits (Gross 1996
; Hazel et al. 2004; Gabriel et al. 2005). In contrast, many studies have found that the use and success of alternative mating tactics are associated with morphological variation between groups of males using different tactics (Shuster 1987; Emlen 1997; Moczek and Emlen 2000; Plaistow and Tsubaki 2000; Heckel and Von Helversen 2002; Forslund 2003). Demographic and social features such as density, sex ratios, and tactic frequencies are also known to be influential (Radwan 1993; Sinervo and Lively 1996; Shuster and Sassaman 1997; Tomkins and Brown 2004). In those few systems where the role of environmental heterogeneity in alternative tactics has been examined, findings suggest that studying alternative mating tactics without taking environmental heterogeneity into consideration could lead to false conclusions about how tactics are used and about the fitness of the tactics (Moczek and Emlen 1999; Gamble et al. 2003; Kolluru and Grether 2005; Lukasik et al. 2006). For example, in bulb mites, a structurally more complex environment resulted in poorer survival for scrambler males, resulting in a switch point at a greater frequency of fighters than in simple environments (Lukasik et al. 2006).
Males of the damselfly Protoneura amatoria exhibit 2 mating tactics within a conditional strategy (Larison B, submitted manuscript). When an individual male uses the perching tactic, he sits high in the vegetation surrounding the mating site. If an unmated female passes nearby, the perching male will attempt to intercept her. When a male uses the hovering tactic, he hovers in a small area just over the surface of the stream and harasses pairs and females that are ovipositing in debris on the water surface. Occasionally, males that use the hovering tactic attempt to physically wrest females from their current mates. Besides the possibility of direct theft of females, this harassment may cause pairs to break up sooner than they otherwise would (Larison B, submitted manuscript), thus increasing hovering males' access to females. In most species of odonates, only one of these tactics is seen. Males of some species defend perching territories, whereas those of other species patrol or hover over the stream (Corbet 1999). The use of perching and flying as alternative tactics in P. amatoria makes it a good species for studying the trade-offs between these 2 common tactics.
Protoneura amatoria experience a spatially and temporally fine-grained environment that is expected to give rise to tactics that are conditional and even reversible (Levins 1962, 1963; Lively 1986; Scheiner 1993; Via 1994). The primary objective of this study is to examine the influence of variation in the light environment on tactic frequencies in this damselfly. Because the use of the hovering tactic is expected to be energetically costly (Dudley 2000), P. amatoria males should be sensitive to environmental conditions that make flight more or less favorable. Thermal irradiance from light can influence flight performance (Josephson 1981; May 1981; Marden 1995; Lehmann 1999), and light conditions may influence visibility and therefore the risk or success of the tactics (Endler 1987; Reynolds et al. 1993; Gamble et al. 2003). Therefore, it is predicted that the frequency of the 2 tactics will be influenced by light conditions.
Population density can also influence the frequency of alternative tactics (Tomkins and Brown 2004); therefore, as a secondary objective, this study examines both the density of unmated males (as a measure of competition among males) and the density of ovipositing females. Ovipositing females appear to be a resource for P. amatoria males using the hovering tactic. As males of many species are known to alter their behavior in response to the availability of females (Carroll 1993; Carroll and Corneli 1995; Eggert and Guyetant 2003; Mills and Reynolds 2003), it is predicted that tactic frequencies will vary in response to variation in the availability of ovipositing females. Ovipositing females are also expected to respond to light for many of the same reasons males should and should also be influenced by the availability of oviposition substrate. Therefore, the physical environment may have important indirect as well as direct influences on alternative tactics, a possibility that has rarely been examined (Rylands 1996).
In this paper, I examine how tactic use in the damselfly P. amatoria varies in response to the light environment, both directly and indirectly through the influence of light on the distribution of ovipositing females. Field observations and experiments were conducted to sort out direct and indirect effects of light and ovipositing females.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONFUNDINGREFERENCES |
|---|
Study species
Protoneura amatoria, like many odonates (Corbet 1999
), visit mating sites specifically for the purpose of encountering mates. Foraging occurs opportunistically and is rarely observed. Therefore, at the mating site, unmated males are pursuing one of 2 tactics, perching or hovering, as described above, and are not engaged in other activities. When a male is using the perching tactic, he may also fly about in the vegetation or well above the stream surface but does not hover immediately over the stream. Males pursuing the hovering tactic may briefly stop hovering and perch, usually in vegetation near the water surface. No courtship has been noted in this species. Once a male captures a female, the pair spends several minutes in copulation (rarely seen) after which the pair flies down to the stream surface. The male then guards the female, remaining in tandem with her while she oviposits on debris floating on the stream surface. Females are most commonly in tandem with males when ovipositing. For brevity, I use the term "ovipositing females" to refer to all females observed ovipositing regardless of whether they are doing so solo or in tandem.
Males using the perching tactic tend to access females that are currently unmated (but not necessarily unmated previously), whereas hovering males are attempting to access females that are currently mated. Although nothing is known about fertilization and postcopulatory selection in Protoneurids specifically, odonate females are known to store sperm and to fertilize their eggs only as they oviposit them (Waage 1979; Corbet 1999). Males of many species are able to remove a predecessor's sperm (Waage 1979, 1986), which typically results in a degree of last male sperm precedence (Siva-Jothy and Tsubaki 1994; Cooper et al. 1996; Hooper and Siva-Jothy 1996).
Observations
The observational part of the study was conducted on 3 streams within Soberania National Park, Panama: Rio Frijolito and Quebrada Juan Grande, both along Pipeline road (approximately 09°09'N, 79°44'W and 09°08'N, 79°43'W), and Rio Sardinilla (approximately 09°05'N, 79°40'W). I collected observational data with the help of assistants between early February and late April in 2004 and 2005. I selected 15 sites on each of the 3 streams. Sites were selected haphazardly on Rio Frijolito and at random on the 2 other streams. Sites were quiet pools distributed along a 2- to 3-km stretch of each stream and represented a range of environmental conditions. The mean pairwise distance between sites was 742 m (range 13–2987 m).
The pairwise distances between some sites in this study are within the range of movement of individual damselflies, and thus, it was important to evaluate whether sites could be treated as statistically independent in the analysis. Based on a study of individually marked males (Larison B, submitted manuscript), I estimate that the probability that a given male was observed in 2 different sites over the course of my observations was 0.13 for the closest sites (those 25 m apart or less) and 0–0.01 for more distant sites (>600 m), with an average percentage of overlap of less than 0.01. I therefore believe it is justified to treat sites as statistically independent. Additionally, analyses were rerun with only those sites that had no other site nearer than 100 m (where the probability of overlap drops to 0.02), and the results are nearly identical to results presented here using the full data set.
I observed each site for 3 days (with the exception of 2 sites that I observed on only 1 day), 4 times a day between the hours of 1000 and 1400. Damselflies begin to show up at the sites as early as 0830, but my observations began at 1000 because preliminary data showed that mating and hovering activity is very low until 1000 (Larison B, unpublished data). I used scan sampling (Martin and Bateson 1999) to collect data on the number of individuals, their sex, and their behavior. I recorded whether unmated males were using the perching tactic (perched or flying near vegetation) or hovering over the surface. Pairs were noted as being in tandem, copulating, or ovipositing. I noted whether unmated females were perching or ovipositing. Because individuals were not marked in this study, it was not always possible to distinguish between resting males that were pursuing the hovering tactic and males that were pursuing a perching tactic. Therefore, data on proportions of males hovering may reflect both tactic switching and bouts of resting on the parts of some males.
I examined several variables relating to the physical environment and 2 relating to the social environment. Physical variables included mean surface illumination (the percentage of the surface of the site typically illuminated at the time of the observation), current weather conditions (whether sunny or cloudy), and canopy cover. Percentage of surface illumination was estimated by eye during each survey period and averaged across those observations made during sunny periods to obtain the mean percentage of surface illumination. This allowed the influence of canopy shade and that of weather to be examined independently. If the sun shone at least briefly during the survey (long enough to carefully estimate surface illumination), I considered the survey period as sunny and scored it as 1, and if it was cloudy for the whole survey, I scored it as 0. At each site, I measured canopy cover, using a handheld densiometer, and the area of the pool. Females oviposit in debris floating on the water surface. I therefore estimated the availability of this oviposition substrate by using a 0.25-m2 quadrat divided into squares of 0.01 m2 and counting the number of squares containing substrate suitable for oviposition. The social factors quantified included male density, as a measure of male–male competition, and the density of ovipositing females.
Experiment 1: response to light
This experiment was conducted in January and February 2005. The purpose of this experiment was to examine the response of ovipositing females and hovering males to light conditions. I selected 16 sunspots along Rio Frijolito for the experiment. Sunspots were selected if they existed for at least 1 h between the hours of 1000 and 1400, were approximately 4 m2, and were at least 50 m from any other experimental plot.
I conducted the experiment in 2 parts in order to examine: A) the response of ovipositing females to light conditions and B) the response of hovering males to light conditions. Sunspots received each of the following 3 treatments in random order: sunny (uncovered), shady (covered with a dark tarp), or control (covered with a clear tarp). I used oviposition substrate to attract ovipositing females. To examine the response of ovipositing females to light conditions, I situated the oviposition substrate within the treatment area so that lighting conditions on the substrate varied with each treatment (Figure 1, rectangles labeled A). The number of ovipositing females was counted within the treatment area. To examine the response of hovering males to light conditions, I placed the oviposition substrate just outside the treatment area (Figure 1, rectangles labeled B). This part of the experiment was designed to change the light conditions in the treatment area without changing the light conditions on the oviposition substrate. This ensured that the number of ovipositing females in the vicinity did not vary in response to the different experimental treatments and therefore eliminated the possibility that variation in the number of ovipositing females would be a confounding factor (see Results). The number of hovering males was recorded within the treatment area, and the number of ovipositing females was recorded in both the treatment area and a meter-wide border around the treatment area where the oviposition substrate was located.
|
In each experiment, the treatment area was 9' x 7' (2.74 x 2.13 m). Tarps were suspended approximately 5' above the water surface. At the beginning of the experiment, treatments were fully set up the day before observations began. Later in the experiment, tarps were removed each afternoon and replaced in the morning at least 1 h before observations. This change was necessary due to tarp theft. Observations were conducted until a minimum of five 2-min observation periods were obtained during full sun conditions. Observations within a treatment were separated by at least 5 min and were treated as subsamples. Observations for each were usually completed within 1 h, which was the amount of time most sunspots persisted. Occasionally, observations for a treatment required 2 days to complete.
Experiment 2: male response to density of ovipositing females
The purpose of this experiment was to examine 1) the response of females to oviposition substrate and 2) whether males change tactics directly in response to the density of ovipositing females. If ovipositing females represent a potential resource for hovering males, then they should respond by increasing hovering when females are abundant and by decreasing it when they are few or absent. A goal of this experiment was to change densities of ovipositing females while not changing densities of unmated males.
I conducted this experiment at 15 sites along Rio Frijolito in March and April 2004. Once sites had been observed in their unmanipulated state, I either cleared away all oviposition substrate at a site or augmented it by adding floating debris from another site. At augmented sites, two 1-m2 floating rectangles (made of polyvinyl chloride tubing and screen) were filled with added substrate and additional debris was scattered at random. Natural levels of substrate cover ranged from 4% to 44%, whereas levels under the removal treatment were between 2% and 8% and levels under the addition treatment were from 22% to 56%. Observations were begun the following day (as above, see Observations). Waiting until the next day avoided direct disturbance due to heavy activity, and making observations soon after the manipulations helped ensure that male densities did not have time to change in response to the treatments. After observations were made on the first set of manipulations, each site received the reverse treatment. I conducted 3 days of observations under each treatment.
Statistics
I analyzed both observational and experimental data using multivariate negative binomial regression (Long and Freese 2001) in Stata 9 (Stata 2005). Negative binomial regression is appropriate for count data with inflated variances. Most of the data did have inflated variances, and where they did not, the model collapsed into a basic Poisson regression. Although my data are count data, the variables of interest are not counts but proportions and densities. I therefore added exposure terms to the regressions to adjust count data relative to other parameters. Where I was interested in the proportion of males using the hovering tactic, I used the total number of unmated males as the exposure term. Where I was interested in the density of ovipositing females, the exposure term was pool size. Because both temporal and spatial variation were of interest, raw data were used in the analyses rather than using site averages. To adjust the degrees of freedom (df) appropriately and to take into account within-site correlation, observations were clustered by site in the analyses. I report the numbers of both observations and clusters for each analysis. I first analyzed data bivariately and included in the full model all variables and interaction terms that were significant to 0.15 or less (Hosmer and Lemeshow 1989). The resulting model remained as the final model.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONFUNDINGREFERENCES |
|---|
Observations
Use of the hovering tactic ranged from 0% (all males perching) to 80% (mean 11%, standard deviation [SD] 14%). The physical and social environments were variable: canopy cover 38–100%, surface illumination 0–95%, area with oviposition substrate 0.03–0.56%, density of ovipositing females 0–0.25 per m2, and density of unmated males 0–0.72 per m2.
Tactic use was associated with several environmental variables, including stream, time of day, mean surface illumination, and whether current weather conditions were sunny or cloudy (Table 1). The proportion of males hovering increased throughout the day and as surface illumination increased, and more males hovered under sunny than cloudy conditions. In preliminary bivariate tests, tactic use was not correlated with canopy cover (negative binomial regression coefficient = 0.003, df = 1, P > 0.704) or with the availability of oviposition substrate (coefficient = 0.582, df = 1, P > 0.370), and I did not enter these variables into the multivariate model.
|
Tactic use was also associated with the density of ovipositing females (Table 1). The proportion of males hovering increased as the density of ovipositing females increased. There was a significant interaction between female density and weather in predicting tactic use. Under sunny conditions, some males hovered even when females were absent, whereas they did not hover in the absence of females under cloudy conditions (Figure 2). Because of this, there was a steeper positive relationship between hovering and female density when it was cloudy (190% increase in hovering per SD change in density of ovipositing females) than when it was sunny (21.9% increase in hovering).
|
The density of unmated males was negatively associated with the proportion of males using the hovering tactic in the multivariate model (Table 1). However, in the preliminary bivariate test, the density of unmated males was positively correlated with the proportion of males using the hovering tactic. In a stream-by-stream examination, the density of unmated males was not a significant factor in any single stream (Rio Frijolito coefficient = 1.1, df = 1, P = 0.75; Juan Grande coefficient = 0.375, df = 1, P = 0.80; and Sardinilla coefficient = 0.322, df = 1, P = 0.43). The density of unmated males is significantly correlated with a number of factors in the model, including the percentage area of substrate (P = 0.002) and the density of ovipositors (P < 0.0001). Qualitatively, the density of unmated males goes up as the proportion of hovering males goes up across streams (Frijolito: density = 0.06, hovering = 0.100; Juan Grande: density = 0.13, hovering = 0.104; Sardinilla: density = 0.19, hovering = 0.13), which may explain why the coefficient changes sign between the bivariate test and the multivariate model. One possible explanation is that male density is acting as a suppressor variable (Pedhazur 1997
); however, its inclusion in or exclusion from the multivariate model has no appreciable impact on the other variables in the model. Because the density of ovipositing females was a significant predictor of tactic use, I also investigated what physical factors were related to the distribution of ovipositing females in order to determine whether the physical environment also had an indirect influence on tactic use. The density of ovipositing females was influenced by oviposition substrate and weather (sunny or cloudy) (Table 2). Like males, females tended to cease activity when clouds blocked the sun. Neither surface illumination (negative binomial regression coefficient = 0.006, df = 1, P = 0.23), canopy cover (coefficient = 0.013, df = 1, P = 0.23), nor time of day (coefficient = 0.688, df = 1, P = 0.65) influenced density of ovipositing females, and none of these variables were entered into the multivariate model.
|
Experiment 1: response to light
Both ovipositing females and hovering males responded to a change in light environment. In experiment 1A, which was designed to test the response of ovipositing females to light conditions (see Figure 1), there were significantly more ovipositing females in the sunny (no tarp) and control (clear tarp) treatments than in the shade (dark tarp) treatment (
2 = 16.89, clusters = 16, observations = 278, df = 2, P = 0.0002; Figure 3A). In experiment 1B, which was designed to test the response of hovering males to light conditions, males hovered significantly more in the sunny and control treatments than in the shady treatment (2 = 2469.67, clusters = 16, observations = 278, df = 2, P = 0.0001; Figure 3B). The number of ovipositing females in the vicinity did not differ between treatments in experiment 1B (2 = 0.062, df = 2, P = 0.73), and therefore, the change among treatments in the number of males hovering is attributable directly to the change in light conditions.
|
Experiment 2: male response to density of ovipositing females
The density of ovipositing females was significantly influenced by the experimental increases and reductions of oviposition substrate, with densities being lowest in the removal treatment, significantly higher in the normal treatment, and higher again in the addition treatment (
2 = 57.10, clusters= 15, observations = 563, df = 2, P < 0.0001; Figure 4A). Sites were successfully manipulated without causing changes in the density of unmated males (2 = 3.01, clusters= 15, observations = 551, df = 2, P = 0.22; Figure 4B). The proportion of males hovering was significantly lower in the removal treatments than in either the normal or the addition treatments (2 = 12.2, clusters= 15, observations = 563, df = 2, P = 0.002; Figure 4C). The proportion of males hovering was slightly, but not significantly, greater in the addition treatment when compared with normal conditions.
|
Because the number of ovipositing females could not be manipulated directly, it is possible that males responded behaviorally to the manipulation of oviposition substrate rather than to the changes in the density of ovipositing females. The observational results, however, suggest that males do not respond to variation in oviposition substrate, and therefore, the response observed in the experiment is likely to be a response to changes in the density of females. An experiment manipulating substrate and females separately would be required to test this inference definitively.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONFUNDINGREFERENCES |
|---|
Multivariate analyses and experimental data show that both male and female P. amatoria respond to physical factors. More males used the hovering tactic as surface illumination increased and when the weather was sunny. Ovipositing females also responded to light, increasing their activity when it was sunny rather than cloudy, but they did not respond to differences in percentage of surface illumination. In addition, ovipositing females reached higher densities at sites with more oviposition substrate. The lack of response to surface illumination by females may be attributable to the fact that ovipositing females tend to cluster wherever oviposition substrate is and, in contrast to males, are not territorial. Because males appear to change tactic use in response to densities of ovipositing females, light conditions and the availability of oviposition substrate may indirectly shape male behavior because of their influence on female behavior and distribution. These findings are concordant with other studies that show a change in male tactic use in response to the light environment (Reynolds et al. 1993
) and studies that show that tactic frequencies are responsive to influences of the physical environment on fitness (Gross 1991; Lukasik et al. 2006).
For P. amatoria, light conditions appear to both cue and constrain male behavior. In sunny conditions, some males hovered even when females were absent. This may be because sun provides an indirect cue that ovipositing females could be present, encouraging males to hover, whereas under cloudy conditions, males pay more direct attention to the presence or absence of females. Other research has found male tactics to be conditional upon availability of females (Carroll 1993; Carroll and Corneli 1995; Eggert and Guyetant 2003; Mills and Reynolds 2003); hence, it follows that they would respond to both females and any cues to the presence of females. It is unclear whether the density of unmated males has an influence on tactic frequencies. Experimental manipulation of male densities is needed to clarify this.
Males hover less under cloudy conditions, even when females are present, indicating that cloudy conditions constrain hovering activity to some degree. Hovering may be constrained under cloudy conditions either due to flight dynamics or due to visual impairment. Individuals in direct sun can increase their thoracic temperature to an average of 5 °C over that of the environment due to irradiative heating (Larison 2007). In insects, warmer individuals are typically able to generate more lift and power (Josephson 1981; May 1981; Coelho 1991; Marden 1995; Lehmann 1999), which leads to greater acceleration and speed (Marden 1989) and therefore greater maneuverability (Marden 1989; Marden and Chai 1991; Kemp et al. 2006). Odonates have high visual acuity in full light, but, in at least some species, vision is slow to adapt to low-light conditions and is relatively poorer in low-light conditions once fully adapted (Autrum and Kolb 1972; Warrant and Pinter 1990). Therefore, it is possible that under cloudy or shady conditions, P. amatoria males are more susceptible to predation and other risks. Predation risk might also be higher in shady spots if fish are more abundant in these areas or able to see damselflies better due to differences in visual properties of lighted and unlighted water (Jagger and Muntz 1993). The influence of light on predation risk and the use of alternative tactics have been shown in other systems (Endler 1987; Reynolds et al. 1993; Gamble et al. 2003), but the possible influences of predation have yet to be tested in this system.
I have shown that P. amatoria males alter tactic use in response to variation in the light environment and that the physical environment may have not only a direct influence on tactic use but also an indirect influence. This indirect influence arises because male tactic use may be responsive to the density of ovipositing females, which can itself be explained by environmental variation. How the environment influences the social and intrinsic parameters on which individuals base their choice of tactic is rarely explored although it is critical to fully understand conditional strategies (Rylands 1996). That males are sensitive to these factors suggests that they are responding to the probable success and risk of each tactic (Reynolds et al. 1993; Moses and Sih 1998; Lukasik et al. 2006) and that the hovering tactic is more fit under high-light conditions and when ovipositing females are present.
It is likely that environmental variation has influenced the evolution of the tactics observed in this species (Stephens 1987; Gabriel 1999; Hazel et al. 2004). However, to fully understand the importance of environmental heterogeneity in shaping tactic use in P. amatoria, the fitness of these tactics under different environmental conditions must be examined. To date, only one study has examined the influence of a physical factor on the fitness of alternative tactics (Lukasik et al. 2006). Future work will address the fitness of each tactic under the physical and social conditions that influence tactic use in P. amatoria and will investigate the influence of morphology as well as how interactions between morphology and environment influence tactic use and fitness.
|
FUNDING |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONFUNDINGREFERENCES |
|---|
The Smithsonian Tropical Research Institute; University of California, Los Angeles; National Science Foundation (NSF-DDIG 0309426).
|
ACKNOWLEDGEMENTS |
|---|
This work could not have been done without the help of my assistants Megan Kefauver, Paulino Gonzalez, Caroline Collins, Sebastian Padrón, and Dennis Anye. I greatly appreciate the guidance provided by Greg Grether, Dan Blumstein, Hugh Dingle, and Peter Nonacs. The manuscript benefited immensely from comments by Judy Stamps, Will Cresswell, and 2 anonymous reviewers.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONFUNDINGREFERENCES |
|---|
Autrum H, Kolb G. The dark adaptation in single visual cells of the compound eye of Aeschna-cyanea. J Comp Physiol (1972) 79:213–232.[CrossRef]
Carroll SP. Divergence in male mating tactics between two populations of the soapberry bug: I. Guarding versus nonguarding. Behav Ecol (1993) 4:156–164.
Carroll SP, Corneli PS. Divergence in male mating tactics between two populations of the soapberry bug: II. Genetic change and the evolution of a plastic reaction norm in a variable social environment. Behav Ecol (1995) 6:46–56.
Coelho JR. The effect of thorax temperature on force production during tethered flight in honeybee Apis-mellifera drones workers and queen. Physiol Zool (1991) 64:823–835.
Cooper G, Miller PL, Holland PWH. Molecular genetic analysis of sperm competition in the damselfly Ischnura elegans (Vander Linden). Proc R Soc Lond B Biol Sci (1996) 263:1343–1349.[Medline]
Corbet PS. Dragonflies: behavior and ecology of Odonata. (1999) Ithaca (NY): Comstock Publishing Associates.
Dudley R. The biomechanics of insect flight: form, function, evolution. (2000) Princeton (NJ): Princeton University Press.
Eggert C, Guyetant R. Reproductive behaviour of spadefoot toads (Pelobates fuscus): daily sex ratios and males' tactics, ages, and physical condition. Can J Zool (2003) 81:46–51.[CrossRef]
Emlen DJ. Alternative reproductive tactics and male-dimorphism in the horned beetle Onthophagus acuminatus (Coleoptera: Scarabaeidae). Behav Ecol Sociobiol (1997) 41:335–341.[CrossRef][Web of Science]
Endler JA. Predation, light intensity and courtship behaviour in Poecilia reticulata (Pisces: Poeciliidae). Anim Behav (1987) 35:1376–1385.[CrossRef][Web of Science]
Forslund P. An experimental investigation into status-dependent male dimorphism in the European earwig, Forficula auricularia. Anim Behav (2003) 65:309–316.[CrossRef][Web of Science]
Gabriel W. Evolution of reversible plastic responses: inducible defenses and environmental tolerance. The ecology and evolution of inducible defenses. Tollrian R, Harvell CD, eds. (1999) Princeton (NJ): Princeton University Press. 286–305.
Gabriel W, Luttbeg B, Sih A, Tollrian R. Environmental tolerance, heterogeneity, and the evolution of reversible plastic responses. Am Nat (2005) 166:339–353.[CrossRef][Web of Science][Medline]
Gamble S, Lindholm AK, Endler JA, Brooks R. Environmental variation and the maintenance of polymorphism: the effect of ambient light spectrum on mating behaviour and sexual selection in guppies. Ecol Lett (2003) 6:463–472.[CrossRef][Web of Science]
Gross MR. Evolution of alternative reproductive strategies: frequency-dependent sexual selection in male bluegill sunfish. Philos Trans R Soc Lond B Biol Sci (1991) 332:59–66.
Gross MR. Alternative reproductive strategies and tactics: diversity within sexes. Trends Ecol Evol (1996) 11:92–98.[CrossRef]
Hazel W, Smock R, Lively CM. The ecological genetics of conditional strategies. Am Nat (2004) 163:888–900.[CrossRef][Web of Science][Medline]
Heckel G, Von Helversen O. Male tactics and reproductive success in the harem polygynous bat Saccopteryx bilineata. Behav Ecol (2002) 13:750–756.
Hooper RE, Siva-Jothy MT. Last male sperm precedence in a damselfly demonstrated by RAPD profiling. Mol Ecol (1996) 5:449–452.[CrossRef][Medline]
Hosmer DW, Lemeshow S. Applied logistic regression. (1989) New York: John Wiley & Sons.
Jagger WS, Muntz WR. Aquatic vision and the modulation transfer properties of unlighted and diffusely lighted natural waters. Vision Res (1993) 33:1755–1763.[CrossRef][Web of Science][Medline]
Josephson RK. Temperature and the mechanical performance of insect muscle. Insect thermoregulation. Heinrich B, ed. (1981) New York: John Wiley & Sons. 20–41.
Kemp DJ, Wiklund C, Van Dyck H. Contest behaviour in the speckled wood butterfly (Pararge aegeria): seasonal phenotypic plasticity and the functional significance of flight performance. Behav Ecol Sociobiol (2006) 59:403–411.[CrossRef][Web of Science]
Kolluru GR, Grether GF. The effects of resource availability on alternative mating tactics in guppies (Poecilia reticulata). Behav Ecol (2005) 16:294–300.
Larison B. Environmental heterogeneity and the evolution of alternative mating tactics in the damselfly, Protoneura amatoria [dissertation]. (2007) [Los Angeles (CA)]: University of California. 81.
Lehmann F-O. Ambient temperature affects free flight performance in the fruit fly Drosophila melanogaster. J Comp Physiol B (1999) 169:165–171.[CrossRef][Medline]
Levins R. Theory of fitness in a heterogeneous environment. I. The fitness function. In: Am Nat (1962) 96:361–373.[CrossRef][Web of Science]
Levins R. Theory of fitness in a heterogeneous environment. II. Developmental flexibility and niche selection. In: Am Nat (1963) 97:75–90.[CrossRef][Web of Science]
Lively CM. Canalization vs. developmental conversion in a spatially variable environment. Am Nat (1986) 128:561–572.[CrossRef][Web of Science]
Long JS, Freese J. Regression models for categorical dependent variables using Stata. (2001) College Station (TX): Stata Press.
Lukasik P, Radwan J, Tomkins JL. Structural complexity of the environment affects the survival of alternative male reproductive tactics. Evolution (2006) 60:399–403.[Web of Science][Medline]
Marden JH. Bodybuilding dragonflies: costs and benefits of maximizing flight muscle. Physiol Zool (1989) 62:505–521.
Marden JH. Large-scale changes in thermal sensitivity of flight performance during adult maturation in a dragonfly. J Exp Biol (1995) 198:2095–2102.[Web of Science][Medline]
Marden JH, Chai P. Aerial predation and butterfly design: how palatability, mimicry and the need for evasive flight constrain mass allocation. Am Nat (1991) 138:15–36.[CrossRef][Web of Science]
Martin P, Bateson P. Measuring behavior. An introductory guide. (1999) 2nd ed. Cambridge (UK): Cambridge University Press.
May ML. Wingstroke frequency of dragonflies (Odonata: Anisoptera) in relation of temperature and body size. J Comp Physiol B (1981) 144:229–240.[CrossRef]
Mills SC, Reynolds JD. Operational sex ratio and alternative reproductive behaviours in the European bitterling, Rhodeus sericeus. Behav Ecol Sociobiol (2003) 54:98–104.[Web of Science]
Moczek AP, Emlen DJ. Proximate determination of male horn dimorphism in the beetle Onthophagus taurus (Coleoptera: Scarabaeidae). J Evol Biol (1999) 12:27–37.[CrossRef][Web of Science]
Moczek AP, Emlen DJ. Male horn dimorphism in the scarab beetle, Onthophagus taurus: do alternative reproductive tactics favour alternative phenotypes? Anim Behav (2000) 59:459–466.[CrossRef][Web of Science][Medline]
Moses JL, Sih A. Effects of predation risk and food availability on the activity, habitat use, feeding behavior and mating behavior of a pond water strider, Gerris marginatus (Hemiptera). Ethology (1998) 104:661–669.[Web of Science]
Pedhazur EJ. Multiple regression in behavioral research: explanation and prediction. (1997) 3rd ed. Fort Worth (TX): Harcourt Brace College Publishers.
Plaistow SJ, Tsubaki Y. A selective trade-off for territoriality and non-territoriality in the polymorphic damselfly Mnais costalis. Proc R Soc Lond B Biol Sci (2000) 267:969–975.[Medline]
Radwan J. The adaptive significance of male polymorphism in the acarid mite Caloglyphus berlesei. Behav Ecol Sociobiol (1993) 33:201–208.[CrossRef][Web of Science]
Reynolds JD, Gross MR, Coombs MJ. Environmental conditions and male morphology determine alternative mating behaviour in Trinidadian guppies. Anim Behav (1993) 45:145–152.[CrossRef][Web of Science]
Rylands AB. Habitat and the evolution of social and reproductive behavior in Callitrichidae. Am J Primatol (1996) 38:5–18.[CrossRef][Web of Science]
Scheiner SM. Genetics and evolution of phenotypic plasticity. Annu Rev Ecol Syst (1993) 24:35–68.[CrossRef][Web of Science]
Shuster SM. Alternative reproductive behaviors: three discrete male morphs in Paracerceis sculpta, an intertidal isopod from the northern Gulf of California. J Crust Biol (1987) 7:318–327.[CrossRef][Web of Science]
Shuster SM, Sassaman C. Genetic interaction between male mating strategy and sex ratio in a marine isopod. Nature (1997) 388:373–377.[CrossRef][Web of Science]
Sinervo B, Lively CM. The rock-paper-scissors game and the evolution of alternative male strategies. Nature (1996) 380:240–243.[CrossRef][Web of Science]
Siva-Jothy MT, Tsubaki Y. Sperm competition and sperm precedence in the dragonfly Nanophya pygmaea. Physiol Entomol (1994) 19:363–366.[CrossRef]
Stata. Stata statistical software, Release 9. (2005) College Station (TX): StataCorp LP.
Stephens DW. On economically tracking a variable environment. Theor Popul Biol (1987) 32:15–25.[CrossRef][Web of Science]
Tomkins JL, Brown GS. Population density drives the local evolution of a threshold dimorphism. Nature (2004) 431:1099–1103.[CrossRef][Medline]
Via S. The evolution of phenotypic plasticity: what do we really know? Ecological genetics. Real LA, ed. (1994) Princeton (NJ): Princeton University Press. 35–57.
Waage JK. Dual function of the damselfly penis: sperm removal and transfer. Science (1979) 203:916–918.
Waage JK. Evidence for widespread sperm displacement ability among Zygoptera (Odonata) and the means for predicting its presence. Biol J Linn Soc (1986) 28:285–300.[CrossRef][Web of Science]
Warrant EJ, Pinter RB. Changes of acuity during light and dark adaptation in the dragonfly compound eye. Z Naturforsch C (1990) 45:137–141.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||