Sex and social costs of escaping in the striped plateau lizard Sceloporus virgatus

Abstract

As a predator approaches, prey base decisions about when to flee on a balance between degree of predation risk and costs of escaping. Lost opportunities to perform activities that may increase fitness are major escape costs. Paramount among these are chances to increase fitness by courting and mating and by driving away sexual rivals. Because sexual selection imposes different social demands on the sexes, social opportunities can have different consequences for males and females, but effects of sex differences in social opportunity costs on escape behavior are unknown. We conducted a field experiment showing that male striped plateau lizards (Sceloporus virgatus) given the opportunity to court or perform aggressive behavior permit closer approach before fleeing, but females do not. Males allowed a simulated predator to approach closer before initiating escape if a tethered male or female rather than a control stimulus was introduced to them, but females initiated escape at similar distances in all conditions. For males, a trade-off between the greater predation risk accepted before fleeing due to the likelihood of enhancing fitness by sexual or aggressive behavior accounts for closer approach allowed in the presence of conspecifics. Mating opportunities are not limiting for females in most species and females often have little to gain by interacting aggressively with other females. Therefore, presence of a conspecific male or female may not justify taking greater risk. Results confirm the prediction of optimal escape theory that flight initiation decreases as cost of escaping increases. The sex difference in effect of presence of conspecifics on flight initiation distance is a consequence of the sex difference in costs of escaping.

Prey exposed to predators must assess degree of risk in relation to costs of escaping, especially opportunities to increase fitness by foraging or interacting socially with conspecifics (e.g., Ydenberg and Dill 1986; Cooper 1999, 2000b; Díaz-Uriarte 1999; Martín and López 1999). Some social costs of escape may differ between sexes due to sexual selection (Cooper 1999, 2003; Díaz-Uriarte 1999). Sexual selection frequently places a premium for males on increasing the number of fertilizations obtained and preventing rivals from “stealing” fertilizations; females having a relatively fixed capacity to produce offspring, in many species, need not attempt to court unfamiliar males or fight with other females to augment their reproductive output (Andersson 1994). Thus, although the sexes may in some species be equally wary when alone, the effect of opportunities to interact socially with conspecifics on escape behavior might differ greatly between males and females. Specifically, males may be less wary in social contexts, whereas wariness of females may be unaffected. Although some data show that males are less wary when fleeing requires them to abandon courtship, mate guarding, or agonistic encounters (Cooper 1999; Díaz-Uriarte 1999; Martín and López 1999), the prediction that social opportunities affect wariness of females has not been tested.

Flight initiation distance, the distance between predator and prey when escape begins, is the primary measure of wariness. Escape theory predicts that flight initiation distance increases with degree of predation risk posed by an approaching predator and decreases with cost of escaping (Ydenberg and Dill 1986; Cooper and Frederick forthcoming). Prey making escape decisions based on trade-offs between fitness consequences of fleeing and not fleeing (losing opportunities to enhance fitness) are predicted to maximize fitness by optimal escape theory (Cooper and Frederick forthcoming) or to flee when costs of fleeing or not fleeing are equal (Ydenberg and Dill 1986). The economic view of escape decisions (Ydenberg and Dill 1986; Cooper and Vitt 2002; Cooper and Frederick forthcoming) has greatly improved our ability to predict flight initiation distance.

Trade-offs of flight initiation distance between predation risk and opportunity costs have been demonstrated for foraging and social opportunities. Prey given the opportunity to feed have shorter flight initiation distances than those lacking feeding opportunities (Cooper 2000b; Cooper et al. 2003; Cooper and Pérez-Mellado 2004). If a prey is engaged in courtship or interacting aggressively with a rival that might steal fertilizations, it is expected to permit closer approach before fleeing, maintaining a balance between fitness costs associated with predation risk and benefits of the social encounter (Cooper 1999; Díaz-Uriarte 1999; Martín and López 1999). Males permit closer approach when courting, guarding mates, or fighting (Cooper 1999; Díaz-Uriarte 1999; Martín and López 1999).

But what of females? They may be warier or use different antipredatory strategies than males (Bauwens and Thoen 1981; Cooper et al. 1990; Schwarzkopf and Shine 1992; Martín and López 1999), but such differences do not permit predictions about effects of social costs on flight initiation distance. Nothing is known about escape trade-offs between risk and either courtship or aggression in females or about possible sex differences in such trade-offs. Females of many species have little to gain by fighting with other females or males except perhaps during certain times during the breeding season. Being courted by an unfamiliar male may be beneficial at times (Griffith et al. 2002; Westneat and Stewart 2003), but in many cases, females reject most courting males, indicating that courtship can be minimally beneficial or perhaps detrimental (Trivers 1972; Le Boeuf and Mesnick 1990; Pitnick and Garcia-Gonzalez 2002). In such cases, flight initiation distance by females is predicted to be unaffected by presence of conspecific males or females. In some species and reproductive states, however, females are aggressive to conspecifics of one or both sexes (e.g., Cooper and Crews 1987; Langmore et al. 2002; Taravosh-Lahn and Delville 2004). Courtship opportunities may be rare or subject to competition (Langmore et al. 2002; Mays and Hopper 2004). Flight initiation distance by females is predicted to decrease when sexual or agonistic interaction is beneficial to them.

We studied effects on flight initiation distance (distance between predator and prey when escape begins) of interactions with introduced conspecifics on potential trade-offs between predation risk and social activities. In the territorial striped plateau lizard, Sceloporus virgatus, males court females frequently and fight with conspecific males before and during the breeding season. In early May when males are aggressive toward each other and court females, females are either nonaggressive or weakly aggressive and are sexually nonreceptive (Vinegar 1972; our observations). However, females become aggressive to both sexes in late May (Vinegar 1972). We conducted a field experiment in early May to test predictions that flight initiation distance decreases for males in the presence of introduced lizards of either sex but is unaffected in females when they are minimally aggressive.

MATERIALS AND METHODS

The study was conducted in the Chiricahua Mountains of southeastern Arizona in the first half of May 2006 at an elevation of approximately 1800 m on warm (26.0–29.8 °C), sunny days when lizards were fully active. All lizards tested were on the ground when initially sighted, either in rocky creek beds or open woods. Lizards are abundant at the site, facilitating data collection. Males were distinguishable from females at a distance because males at the time of the study had swollen tail bases, brighter blue throat patches, and few or no dark dorsal spots.

To test escape responses of lizards, a human investigator simulated a predator by approaching them directly (e.g., Martín and López 1996; Cooper et al. 2003; Webb and Blumstein 2005). When a lizard was sighted, the investigator approached it very slowly and placed either a conspecific (experimental condition) tethered by a 0.5-m string to a 1.7-m rod or a tethered stick (control condition) of similar size and color within 0.3 m of the lizard, providing unobstructed views of the tethered lizard and the investigator to the focal lizard. The investigator's arm (0.8 m) was outstretched during stimulus placement, so that the investigator's body did not come closer to the lizard than about 2.5 m. After putting the rod on the ground, the investigator slowly withdrew to 5 m from the focal lizard. After 10 s of immobility, the investigator approached the focal lizard at a very slow, practiced speed (33.7 ± 2.1 m/min, n = 10, data are mean ± 1.0 standard error) until the lizard fled. Approach speed during placement of the stimulus was not measured but was much slower than approach speed during trials. Flight initiation distance was measured to the nearest 0.1 m.

Pseudoreplication was avoided by constantly monitoring each lizard between initially observing it completing collection of its data by moving to a new location far enough away to prevent the previously tested lizard from having moved that far before testing another lizard and by moving through a given area only once. A single investigator wearing similarly colored clothing served as predator in all trials.

Although many lizards in campgrounds in the Coronado National Forest in the Chiricahua Mountains appear to be habituated to human presence, those tested in this study were in areas where human disturbance is infrequent. They readily fled when approached by a person.

Using a human researcher as a simulated predator has the disadvantages that researchers are not natural predators of S. virgatus and must know the experimental treatment during each trial, raising the possibility that unconscious bias might affect results. Typical predators of small lizards include raptors, other predatory birds, mammals, such as coyotes and foxes, and snakes (e.g., Burleigh 1958; Sherbrooke 1990; Conant and Collins 1998; Sherbrooke and Middendorf 2004; Sherbrooke and Mason 2005). Although human beings are not natural predators of S. virgatus, these lizards flee when approached by a person. Experimental tests in which humans served as simulated predators have confirmed predictions of escape theory for numerous factors affecting predation risk and cost of escape (e.g., Burger and Gochfeld 1990; Dill 1990; Bulova 1994; Cooper 1997, 1998, 2000a; Smith 1997). To eliminate possible bias due to knowledge of experimental conditions and groups, we standardized methods of approaching in all conditions, including the approach speed.

Stimulus lizards were tethered by attaching string to tape about the middle of the torso. During trials, focal males sometimes performed aggressive or courtship displays but did not attack the stimulus lizards. Each stimulus lizard was used in 3–4 trials. At the conclusion of testing, the tape was removed from stimulus lizards easily without causing any apparent discomfort or damage to scales. The lizards were released at their sites of capture within 2 h.

We used a repeated-measures design in which each lizard was tested twice, once with a tethered conspecific and once with the control stick. Thus, each individual served as its own control for escape tendency. After the first trial was conducted, the investigator withdrew and observed the lizard until 30 s after it resumed normal activity before conducting the second trial. Order of trials was counterbalanced within individuals and was alternated for sexes of introduced lizards to prevent sequential biases. Focal lizards were tested in the order encountered, but their sexes were interspersed so that no major differences in times of day or data occurred. Because each focal lizard was tested twice, its position when tested differed in the 2 trials. If lizards were closer to refuges during second trials, thus having shorter flight initiation distances, counterbalancing would eliminate effects of such bias. Furthermore, lizards were tested on the ground during both trials and all lizards were close to multiple refuges. Lizards were tested only if no other lizards were detected nearby that might have influenced their behavior.

The experiment had a mixed factorial design with conspecific versus control as a repeated, within-groups factor and sex of the tested (focal) lizard and effect of sameness or difference of sex between the introduced and focal lizard as between-groups factors. Data met assumptions required for repeated-measures analysis of variance (ANOVA) (Zar 1996) using a general linear model. An additional ANOVA was conducted adding trial number as a factor to those in the previous analysis. All statistical tests were 2-tailed with α = 0.05.

RESULTS

Focal males typically responded to tethered conspecifics by performing aggressive or sexual displays; focal females rarely responded overtly to tethered lizards. Neither sex performed any displays in the presence of control stimuli.

Males permitted closer approach before fleeing in the presence of introduced conspecific males or females than control stimuli, but the presence of tethered conspecifics did not affect flight initiation distance by females (Table 1). The main effect of whether the introduced lizard was of the same sex as the focal lizard was not significant (F_1,46 = 2.12, P > 0.10). The main effect of sex of the focal animal was significant, flight initiation distance being greater for focal females than focal males (F_1,46 = 4.15; P = 0.047). The interaction between sameness of sex of the 2 lizards and sex of the focal lizard was not significant (F_1,46 = 1.76; P > 0.10). Flight initiation distances were significantly shorter in the presence of introduced lizards than control stimuli (F_1,46 = 46.66; P < 1 × 10⁻⁶). The interaction between introduction condition and sameness of sex of the 2 lizards was not significant (F_1,46 = 2.11; P > 0.10).

Table 1

Flight initiation distances (m) for tethered introductions to Sceloporus virgatus

Focal lizard	Introduced lizard	Control
Male	Female
$\stackrel{-}{\text{X}}$	0.38	1.16
SE	0.14	0.12
Range	0.1–1.6	0.8–2.0
	Male
$\stackrel{-}{\text{X}}$	0.42	1.09
SE	0.10	0.11
Range	0.2–1.3	0.6–1.9
Female	Female
$\stackrel{-}{\text{X}}$	0.88	0.81
SE	0.13	0.18
Range	0.1–1.7	0.5–2.2
	Male
$\stackrel{-}{\text{X}}$	1.10	1.26
SE	0.14	0.13
Range	0.5–2.0	0.2–2.0

SE, standard error.

Sample sizes were 14 for males introduced to females and 12 for all other data.

The most important effect was a strong interaction between introduction condition and focal sex (F_1,46 = 36.59; P < 1 × 10⁻⁶): flight initiation distance was shorter in males in the presence of introduced lizards of either sex than the control stimulus but in females was not affected by presence of introduced lizards (Figure 1). The 3-way interaction among introduction condition, sameness of sex of the 2 lizards, and sex of the focal lizard was not significant (F_1,46 = 0.32; P > 0.10). For males, the effects of introducing male and female conspecifics on flight initiation distance were nearly identical decreases relative to the control condition (Figure 2).

Figure 1

Flight initiation distance (FID) of adult Sceloporus virgatus. FID was much shorter for males when conspecifics of either sex were introduced than in the control condition but for females was similar for introduced conspecifics and the control stimulus. Error bars are 1.0 standard error.

Figure 2

Flight initiation distances by male Sceloporus virgatus. FID did not differ between sexes of introduced conspecifics (n = 12 for each sex) but were much shorter in trials with conspecifics than with control stimuli. Control data are pooled (n = 24). Error bars are 1.0 standard error.

In the test that included trial number as a third factor, an identical pattern of significance for the factors included in the first analysis was obtained. The main effect of trial sequence was not significant (F = 3.93; P = 0.054), but there was a strong trend for flight initiation distance to be greater in second trials. None of the interactions involving trial number were significant.

DISCUSSION

Significant main effects of tethered lizard versus control stimulus and of focal sex cannot be interpreted simply because their interaction was significant. In a previous study, flight initiation distance did not differ between sexes in S. virgatus that were not adjacent to conspecifics (Smith 1996). Our results are consistent with this finding due to the strong interaction between sex and presence or absence of a conspecific: Flight initiation distance for females was similar to that of males when focal females were alone; females fled at greater distances than males only when conspecifics were present.

Social encounters had markedly different effects on flight initiation distance in males and females in exactly the manner predicted by sex differences in benefits of social interactions. Resident territorial males potentially can obtain substantial gains in fitness by engaging in aggressive behavior to drive away nonresident males that might fertilize eggs of females residing in their territories if allowed to stay there. Resident males also may enhance their fitness by gaining fertilizations through courtship and mating with unfamiliar females. These benefits are consistent with the shorter flight initiation distances by males in the presence of conspecifics than control stimuli.

Flight initiation distances of focal males did not differ between sexes of introduced lizards. This might be thought to indicate that males could not readily distinguish between introduced males and females. However, color patterns of the 2 sexes were sufficiently distinct for human investigators to identify sex at a distance of several meters even though the bright orange reproductive coloration of females was absent during our study. The swollen tail base of males is an even more reliable cue. Furthermore, social displays performed by focal males were appropriate for the sexes of tethered lizards. This suggests that benefits may be approximately equal for courting an unfamiliar female or interacting aggressively with an unfamiliar (intruder) male.

Equality of benefits is plausible because females were presumably not fertilizable when the study was conducted. Females develop orange coloration when their follicle is enlarged just prior to ovulation, providing males not only an additional cue for sex recognition but also an indicator of fertilizability. Males in the lizard Holbrookia propinqua appear to use this cue, courting females intensely when their female coloration appears (Cooper and Crews 1987, 1988; Cooper and Greenberg 1992). Thus, flight initiation distance by males might be shorter for unfamiliar females that exhibit bright reproductive coloration than those that do not.

For females, the impact of social encounters is not always so clear-cut as for males. Even if a female might mate with a male other than the territory holder at some time during the mating season, she is likely to have numerous opportunities to do so in populations in which conspecifics are as abundant as they are at our study site. For a female that would reject courtship in the current encounter with an unfamiliar male, fleeing from the vicinity of the male does not diminish reproductive success. Thus, similarity of flight initiation distance for introduced males and control stimuli is a consequence of the male's temporary presence having little or no impact on the female's fitness.

A similar interpretation pertains to interactions between 2 females. If females compete for important resources such as food or males (Langmore et al. 2002), aggression may allow dominant females to compete more effectively. However, if competition between females is unimportant or restricted to certain seasons, presence of a conspecific female may not induce aggression and cost of fleeing would not be increased. At the time of the study, we observed very little aggression between females, as noted by Vinegar (1972). Flight initiation distances in the experimental conditions presumably were similar because presence of an unfamiliar female did not affect cost of fleeing.

Our findings strongly confirm predictions of a sex difference in the effect of conspecific presence on flight initiation distance based on escape costs in striped plateau lizards in early May. They cannot be generalized to all species, ecological conditions, and reproductive states. Enough must be known about the natural history and social structure of a species to base predictions on social costs. In S. virgatus, females contain oviductal eggs in late May and are often intensely aggressive toward conspecific males and females (Vinegar 1972). Aggressive rejection of courtship may be beneficial for gravid females because courtship and mating may attract the attention of predators at a time (Cooper and Greenberg 1992) when the weight of a clutch of eggs reduces their maximum running speed (Shine 1980; Cooper et al. 1990). The selective impetus for increase in aggression toward conspecific females is not known, but competition for oviposition sites or, subsequently, for food to recover nutrients used in reproduction are among the possibilities. We can predict that during intervals when females are aggressive to conspecifics, flight initiation distance for females decreases in the presence of conspecifics of both sexes. On the other hand, if females are slowed sufficiently by the weight of oviductal eggs when they are aggressive, the optimal flight initiation distance might increase to maintain a margin of safety. Empirical study is needed to determine the existence and relative importance of such effects.

This work was partially supported by a Pippert Science Research Scholar award to W.E.C. and by our respective institutions. It was conducted according to research protocol 00-037-03 of the Purdue Animal Care and Use Committee and in accordance with state and federal laws of the United States.