SEARCH

SEARCH BY CITATION

Keywords:

  • bottom-up;
  • plasticity;
  • rapid evolution;
  • Sceloporus;
  • Solenopsis

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

Responses to novel threats (e.g. invasive species) can involve genetic changes or plastic shifts in phenotype. There is controversy over the relative importance of these processes for species survival of such perturbations, but we are realizing they are not mutually exclusive. Native eastern fence lizards (Sceloporus undulatus) have adapted to top-down predation pressure imposed by the invasive red imported fire ant (Solenopsis invicta) via changes in adult (but not juvenile) lizard antipredator behaviour. Here, we examine the largely ignored, but potentially equally important, bottom-up effect of fire ants as toxic prey for lizards. We test how fire ant consumption (or avoidance) is affected by lifetime (via plasticity) and evolutionary (via natural selection) exposure to fire ants by comparing field-caught and laboratory-reared lizards, respectively, from fire ant-invaded and uninvaded populations. More naive juveniles from invaded populations ate fire ants than did adults, reflecting a natural ontogenetic dietary shift away from ants. Laboratory-reared lizards from the invaded site were less likely to eat fire ants than were those from the uninvaded site, suggesting a potential evolutionary shift in feeding behaviour. Lifetime and evolutionary exposure interacted across ontogeny, however, and field-caught lizards from the invaded site exhibited opposite ontogenetic trends; adults were more likely to eat fire ants than were juveniles. Our results suggest that plastic and evolutionary processes may both play important roles in permitting species survival of novel threats. We further reveal how complex interactions can shape adaptive responses to multimodal impacts imposed by invaders: in our system, fire ants impose stronger bottom-up selection than top-down selection, with each selection regime changing differently across lizard ontogeny.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

Antagonistic interactions impose strong selective pressure on both predators and prey and subsequently play an important role in structuring populations, communities and ecosystems (Crawley, 1992; Langerhans, 2007). There is increasing recognition that evolutionary responses to novel antagonistic interactions caused by environmental change, including the introduction of non-native species, can occur at ecological time scales (Hairston & Dillon, 1990; Hendry & Kinnison, 1999; Mooney & Cleland, 2001; Strauss et al., 2006). Environmental change associated with invasive species can modify mortality rates, behaviour, physiology and development of native species (Case & Bolger, 1991; Suarez & Case, 2002; Wauters et al., 2002; Moore et al., 2004). In producing these ecological effects, invaders alter regimes of natural selection.

Selective pressures can be complex, being generated through top-down or bottom-up effects, for example, and can change throughout ontogeny (e.g. Menge, 1995; Menge et al., 1996; Gagliano et al., 2007). Invaders acting as novel threats at mid-trophic levels are able to simultaneously impose both top-down and bottom-up selective pressures on other species. For instance, the cane toad (Bufo marinus) acts as both a predator and toxic prey within the communities it inhabits, changing selection regimes for many species in these communities (e.g. Greenlees et al., 2006, 2010; Nelson et al., 2011). Impacts of invaders on native populations also may vary across different life-history stages as a consequence of changes in ecological interactions across ontogeny (e.g. Gagliano et al., 2007). As a result, invaders can impose different selective pressures on juveniles vs. adults, not only in strength but also in the direction of trait values (e.g. selection for longer or shorter limbs) and trophic structure (e.g. bottom-up or top-down). For example, juveniles and adults may differ in (i) their propensity to prey on the invader, due to differences in body size (Santos Filho, 1997), (ii) their probability of encountering an opportunity to prey upon an invader and/or succumbing to attack, due to differences in habitat use or behaviour (Mancinelli, 2010; Bos et al., 2011), and/or (iii) their vulnerability to the effects of envenomation, due to the mass-dependent nature of toxin dosage (Read et al., 1978). If the selective pressure imposed by invaders across ontogenetic stages differs from that experienced in the absence of invasion, it may result in a reversal, or loss, of natural ontogenetic shifts in diet and/or behaviour within native populations.

Responses to such novel threats can be driven by natural selection, resulting in genetic changes, or plastic nongenetic shifts in phenotype. There is controversy over the relative importance of these processes for species survival of novel threats (Hoffmeister et al., 2005; Ellner et al., 2011), but we are realizing they are not mutually exclusive (Losos et al., 2004). Recent reviews highlight the critical need to understand the extent to which phenotypic plasticity (including learning) and evolutionary change ameliorate the impact of invasive species, and to assess the consequences of adaptive responses for the persistence of native populations and therefore communities (Hoffmeister et al., 2005; Strauss et al., 2006). It is clear that we need to merge ecological and evolutionary perspectives, by assessing responses to novel pressures during lifetimes and across generations, for example, if we are to obtain a complete picture of how communities are structured.

The invasion of red imported fire ants into eastern fence lizard (Sceloporus undulatus) habitat provides an excellent opportunity to examine the direct effects of trophic interactions across ontogeny (e.g. Suarez & Case, 2002). This globally important invader has been introduced to over seven countries, including the USA in the mid-1930s where it has since spread across fourteen states (Tschinkel, 2006; Code of Federal Regulations, 2012). The small size and social foraging mode of these venomous ants make them effective predators, allowing them to kill organisms much larger than themselves (Garmestani et al., 2004; Smith et al., 2007; Wetterer et al., 2007; Langkilde, 2009; Wilcoxen & Rensel, 2009). The top-down impact of fire ants as novel predators has received much attention (Allen et al., 2004; Tschinkel, 2006); fire ants are able to locate and kill hatchling tortoises (Landers et al., 1980; Epperson & Heise, 2003), nestling birds (Lockley, 1995; Krogh & Schweitzer, 1999) and adult lizards (Langkilde, 2009; Freidenfelds et al., 2012) in the wild. In some instances, the novel selective pressure imposed by this invader can have detrimental population-level effects (Lockley, 1995; Allen et al., 1997, 2001; Buhlman & Coffman, 2001; Dabbert et al., 2002). For example, fire ants appear to have changed the ontogeny of behavioural responses of fence lizards to fire ant attack (Langkilde, 2009, 2010). Within uninvaded sites, the majority of juvenile fence lizards (which are vulnerable to even native nonvenomous ants; Vitt, 2000) perform body-twitches that remove attacking ants, and quickly flee from attack, allowing them to survive these encounters. This responsiveness is largely lost in adult lizards, which are unlikely to be vulnerable to native ants; approximately half of the adults from uninvaded sites remain motionless during fire ant attack, often with lethal consequences (Langkilde, 2009). In contrast, this antipredator behaviour is prevalent throughout ontogeny within fire ant-invaded sites, with approximately 90% of adult lizards exhibiting this neonatal behavioural response to ants.

Lizards (including fence lizards and horned lizards, Phrynosoma coronata) will consume fire ants during attack (and possibly also during regular foraging), which sting them inside the mouth (Webb & Henke, 2003; Boronow & Langkilde, 2010). Resulting envenomation can kill these native ant predators (Langkilde & Freidenfelds, 2010). Red imported fire ants out-compete native ant species that serve as prey for many native taxa (Porter & Savignano, 1990; but see King & Tschinkel, 2008), and the resulting reduction in availability of native ant prey likely increases encounter rates between fire ants and native ant predators. The implications of fire ants as novel toxic prey, however, have been largely ignored (but see Langkilde & Freidenfelds, 2010). There is an ontogenetic shift in fence lizard diet, with ants comprising 80% of juvenile diets but only 50% of adult diets (Demarco & Ferguson, 1985; Parker, 1994). This, together with the mass-dependent effect of venom, whereby smaller animals succumb more easily to envenomation, suggests that the bottom-up impact of fire ants as toxic prey may be important, especially for juveniles. As a result, we expect that there should be selective pressure for juveniles to avoid eating fire ants.

Here, we examine the poorly understood, but likely important, bottom-up effects of invasive fire ants on native lizards. We merge ecological and evolutionary perspectives to examine the relative roles of plasticity and natural selection in population divergence in behaviour by determining how ontogenetic stage and prior exposure to fire ants affect the propensity of native fence lizards to prey upon these venomous invaders. Differences in behavioural response to invaders between invaded and uninvaded populations may reflect (i) plastic responses resulting from differences in lifetime exposure to the invader and/or (ii) intrinsic (e.g. genetic or maternal) population differences resulting from evolutionary exposure (over generations) to the invader. If the predicted difference in ontogenetic selection regime imposed by these venomous prey is occurring, juveniles from uninvaded sites will be more likely than adults to prey upon fire ants, but juveniles from invaded sites will be less likely than adults to prey upon fire ants. By assessing the contributions of lifetime (using field-caught lizards) and evolutionary (using laboratory-reared lizards) exposure to fire ants in driving these patterns (Fig. 1), we shed light on the mechanisms shaping responses to this potential bottom-up selection.

image

Figure 1. Schematic of experimental design. The design includes three factors, the fire ant invasion status of the source populations, the treatments controlling for naivety (laboratory-reared vs. field-caught) and ontogenetic stage (juvenile vs. adult).

Download figure to PowerPoint

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

Study species

We examined fire ant consumption by field-caught and laboratory-reared fence lizards from a population that has been invaded by fire ants for approximately 70 years (Solon Dixon Forestry Education Center, Escambia County, Alabama, 31°09′49″N, 86°42′10″W) and a population that has not yet been invaded by fire ants (St. Francis National Forest, Lee County, Arkansas, 34°43′50″N, 90°42′18″W). These populations are endpoints of a previously identified gradient in antipredator-related morphology and behaviour of lizards in response to fire ants, associated with time since fire ant invasion (Langkilde, 2009). Molecular markers, coloration and scalation, and morphology of museum specimens collected prior to invasion suggest that this species is relatively undifferentiated across its range (Smith et al., 1992; Leache & Reeder, 2002; Miles et al., 2002; Langkilde, 2009; see Langkilde, 2010 for more detail). Habitat used by lizards at these sites is similar: eight habitat characteristics (canopy openness, vegetation cover, leaf litter, rock cover, wood cover, soil cover, diameter of lizard perch and diameter of nearest tree) measured at the points of capture of 40 lizards at each site. Principle component analyses grouped these variables into three axes explaining open canopy/leaf litter, absence of vegetation cover, and large trees/wood cover, and these did not differ between the sites used in this study (sites 1 and 4 in Langkilde, 2009, appendix D). Thus, the major difference between these environments appears to be the presence or absence of the invasive fire ant.

Our multifactor assessment on lizards of lifetime and evolutionary exposure to fire ants through ontogeny consisted of 392 field trials, and husbandry of 225 lizards from egg to adult in addition to 24 gravid females captured and brought back to the laboratory for oviposition. We captured 64 and 68 adults and 21 and 14 juveniles from the invaded and uninvaded population, respectively, using a hand-held noose. Laboratory-reared animals (n = 107 and 118 from the invaded and uninvaded population, respectively) were collected as eggs from females from each of these populations. After hatching, these lizards were housed in groups of 4–6 within enclosures (56 × 40 × 30 cm, L × W × D) lined with paper towelling and furnished with a shelter, water bowl and heat source. Juveniles were housed in mixed-sex groups until maturity, at which time they were housed at the same density within single-sex groups. All lizards were provided water ad libitum and fed crickets (Acheta domesticus) daily (juveniles) or every other day (adults). Approximately half of the laboratory-reared lizards were tested for their feeding response to fire ants at a mean of 2 weeks of age (5–23 days; n = 55 and 66 from the invaded and uninvaded population, respectively); the remaining lizards (n = 52 and 52 from the invaded and uninvaded population, respectively) were tested after they had reached maturity (approximately 1–2 years of age), as determined by snout-vent length (SVL), as well as the presence of developed, ventral badges (at least three-fourths of blue coloration developed and presence of black coloration) for males. Size at maturity was previously determined for a nearby population in Mississippi as 53 mm SVL for males and 58 mm SVL for females based on the presence of enlarged testes and vasa deferentia, and vitellogenic follicles or oviducal eggs, respectively (Parker, 1994). Both males and females were used in behaviour trials. Sex did not significantly affect fire ant consumption, or significantly interact with any of the other factors (all P-values > 0.059), so data for males and females were grouped for analyses.

Field-caught and laboratory-reared lizards from both sites were held in individual breathable cotton bags in coolers kept between approximately 15 and 25 °C for < 48 h following capture from the field or their enclosures. During this time, field-caught lizards from the uninvaded site and all laboratory-reared lizards were transported to the invaded site for testing. Field-caught lizards were then either tested immediately (approximately 88% of lizards from each site) or individually housed in enclosures (30 × 20 × 25 cm, L × W × D), provisioned as described above, at the invaded site for 1 week prior to testing.

Feeding behaviour

We quantified fire ant consumption by lizards during behaviour trials (described in detail in Langkilde, 2009). The behaviour trials were conducted in the field on natural fire ant mounds in order to ensure that the behaviour of the lizards and the fire ants was as natural as possible. Mounds were located in an area away from where lizards were collected, and therefore out of their typical home ranges. Briefly, we tethered each lizard with a 1-m cotton leash attached to a tent peg positioned 40 cm from an active fire ant mound at the invaded site. The purpose of these tethers was to prevent lizards escaping during the trials (critical as we tested lizards away from their source populations) while allowing them to move and behave unimpeded. Tethering in this manner does not affect the behavioural response of these lizards to fire ants (T. Langkilde, unpublished), or other behaviour of this and congeneric species (Vinegar, 1975; Cooper, 2009). Each lizard was positioned facing a mound which had been lightly disturbed with a stick to encourage approximately 6 fire ants to emerge. We tapped the lizard on the base of the tail to encourage it to run onto the mound, using the tether to guide and slow the lizard. Trials commenced when the first fire ant moved onto the stationary lizard, and we recorded the consumption of fire ants by lizards during the trial (see Supporting Information, Movie S1). The number of ants that encountered the lizards during these trials averaged 4 ± 0.15 SE. To ensure that lizards received only sublethal doses of venom from fire ants stings, but had adequate time to eat ants, we removed the lizard from the mound after 30 s (juveniles) or 60 s (adults) if the lizard had not fled the mound, and removed any attacking ants, and no lizards died during these trials. It is unlikely that our rescue had a significant impact on ant consumption because most juveniles fled within 30 s, and lizards that did not flee by the end of the trial would have soon become paralysed and therefore unable to consume ants (Langkilde, 2009). Because fire ants could attack lizards during these trials, as in nature, we tested whether foraging opportunity, measured as the amount of time before a lizard fled attack (latency to flee), affected whether or not a lizard ate fire ants.

Analyses

We tested whether lifetime exposure to fire ants (field-caught lizards from the invaded population vs. the uninvaded population), evolutionary history with fire ants (laboratory-reared lizards from invaded vs. uninvaded source populations) and ontogenetic stage (juvenile vs. adult) affected whether or not lizards would eat fire ants. Specifically, we employed a generalized linear model (GZLM) to examine whether naivety (field-caught vs. laboratory-reared), invasion status (invaded or uninvaded), ontogenetic stage (juvenile or adult) and the interaction of these three factors affected the proportion of individual lizards that consumed fire ants (binary response variable of Yes/No; see Fig. 1). Latency to flee (seconds) was used as a covariate in the model to account for effects of feeding opportunity on fire ant consumption. Significant interaction terms were explored using post hoc GZLMs to parse the significant trends. Maternal (clutch) identity was only known for laboratory-reared lizards and did not significantly affect fire ant consumption (all P-values > 0.4), so clutch information was not included in subsequent statistical models. We also tested for a relationship between whether lizards ate fire ants or not during the trials, and the number of ants encountered, using logistic regression (n = 391). We recorded the number of ants consumed (0–6 per lizard) for lizards in all treatments, but only for a subset of field-caught adult lizards (n = 24 from each site); analysis of these data (n = 308, GZLM using a Poisson distribution for these count data) produced qualitatively similar results to analysis of whether or not lizards ate fire ants described earlier in this section, and we do not present these results here. Among the lizards that did eat fire ants, we tested for correlations between the number of ants eaten, the number of ants that a lizard encountered, and its latency to flee (juveniles, n = 63; adults, n = 34). All statistical analyses were conducted using spss (SPSS Inc., 2006, Chicago, IL, USA).

We estimated rates of divergence in fire ant-eating behaviour between populations to facilitate comparison between invasion status and ontogenetic stages within our study and with other studies. We estimated behavioural divergence between populations in haldanes, which measure rates of change per generation, using estimated marginal means (i.e. means corrected for covariates). Haldanes are a more appropriate measure than darwins for behavioural traits, and more universally comparable for other reasons (see Hendry & Kinnison, 1999). For this study, haldanes were calculated over 36.96 generations based on 1.84 years per generation (Parker, 1994) and 68 years since fire ant invasion of the invaded site (Langkilde, 2009). For field-caught individuals, we considered the divergence to be phenotypic (hp) because genetic and environmental influences are indistinguishable. For laboratory-reared individuals, we considered the divergence to be caused more by genetic changes (hg) than effects of the environment, as animals were raised under common-garden conditions (i.e. Hendry & Kinnison, 1999). We note that rates of divergence are described instead of rates of evolution because our synchronic experimental design does not allow us to measure the rates of evolution as we did not follow specifically the changes in trait values through time. Any evolutionary change may have occurred in just a few generations and then subsided, with trait values persisting under stabilizing selection for an extended period of time (within which we measured the trait values).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

Lifetime exposure to fire ants, evolutionary history with fire ants and ontogenetic stage all affected the proportion of lizards that consumed fire ants, as indicated by a significant 3-way interaction: naivety (field vs. laboratory) × invasion status (invaded vs. uninvaded) × ontogenetic stage (adult vs. juvenile) (inline image = 6.54, P = 0.011). Feeding opportunity, which was accounted for in this model as latency to flee, played a significant role in determining fire ant consumption: the longer a lizard endured an attack, the more likely it was to eat fire ants during this time (latency to flee, inline image = 11.62, P < 0.001). To facilitate the interpretation of the 3-way interaction, we analysed the data for field-caught and laboratory-reared lizards separately, testing for effects of invasion status and ontogenetic stage on fire ant consumption by lizards within these groups. This allowed us to specifically examine the effects of both evolutionary and lifetime exposure to fire ants (behaviour of field-caught lizards) and effects of just evolutionary history with these invaders (intrinsic behaviour of laboratory-reared lizards) on fire ant-eating behaviour across ontogenetic stages.

Within the field-caught lizards, there was a significant interaction between invasion status and ontogenetic stage (inline image = 5.00, P = 0.025). This is driven by the fact that the ontogenetic trend in fire ant-feeding behaviour differs with invasion status: more juveniles than adults from the uninvaded population ate fire ants, whereas the opposite was true for lizards from the invaded population (Fig. 2a). Indeed, more field-caught adults from the invaded population ate ants than did those from the uninvaded population (inline image = 3.90, P = 0.048; Fig. 2a,b). Feeding opportunity (latency to flee from attack) was not a significant predictor of fire ant consumption within these field-caught lizards (inline image < 0.01, P = 0.981), so was not included in the post hoc model. The interaction between lifetime and evolutionary exposure to fire ants is reflected in the rate of phenotypic change between populations (which we estimated in haldanes; see 'Materials and methods'): the rate of divergence in adults is equivalent to hp = 0.0372, and in juveniles it is equivalent to hp = −0.0359. A positive haldane value indicates an increase in the trait value (in this case, an increase in fire ant consumption), and a negative value indicates a decrease (avoidance of fire ants).

image

Figure 2. Fire ant consumption by lizards. The proportion of field-caught (a) and laboratory-reared (b) adult (open squares) and juvenile (solid squares) fence lizards, Sceloporus undulatus, from a fire ant-invaded and uninvaded site that consumed fire ants during fire ant attack. Points represent mean values ± 1 standard error.

Download figure to PowerPoint

Within the laboratory-reared lizards, more juveniles consumed fire ants than did adults (inline image = 24.23, P < 0.001), and more lizards from the uninvaded population consumed fire ants than from the invaded population (inline image = 4.19, P = 0.041; Fig. 2b). There was no significant interaction between these factors (inline image = 0.41, P = 0.524; Fig. 2b), indicating that the difference in feeding behaviour between juvenile and adult laboratory-reared lizards was similar in both populations, but fewer juveniles and adults from the invaded population ate ants than did those from the uninvaded population. Feeding opportunity predicted fire ant consumption by laboratory-reared lizards; lizards that fled sooner from attack were less likely to eat ants (latency to flee, inline image = 17.58, P < 0.001). The effect of evolutionary exposure to fire ants on fire ant consumption by lizards is reflected in the rate of intrinsic change in this behaviour: the rate of divergence in adults is equivalent to hg = −0.0191, and in juveniles it is equivalent to hg = −0.0107.

Whether or not a lizard consumed fire ants during a trial was not affected by the number of ants that it encountered during the trial (adults, β235 = −0.01, P = 0.802; juveniles, β156 = 0.04, P = 0.739). Among lizards that did consume fire ants during a trial, the number of ants that a lizard encountered during the trial did not affect its latency to flee (adults, r34 = 0.11, P = 0.555; juveniles, r63 = 0.18, P = 0.155) or the number of ants it ate (adults, r34 = 0.27, P = 0.125; juveniles, r63 = 0.05, P = 0.682), and the number of ants a lizard ate during a trial was not related to its latency to flee (adults, r34 = 0.15, P = 0.410; juveniles, r63 = 0.11, P = 0.403).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

Our data suggest that a lizard's propensity to eat fire ants is a result of both lifetime and evolutionary exposure to fire ants, with the interaction of these forces across ontogeny reflecting differences in selective pressures imposed on adults vs. juveniles. Within laboratory-reared lizards from both fire ant-invaded and uninvaded populations, more juveniles than adults ate fire ants. This likely reflects the innate ontogenetic shift in diet away from ants in adulthood (likely linked to ecological changes associated with body size; Brooks & Dodson, 1965; Malmquist et al., 1992; Webb & Shine, 1993; Puvanendran et al., 2004). As we predicted, however, there is an intrinsic difference in the proportion of lizards from the invaded vs. uninvaded populations that consume fire ants: fewer laboratory-reared lizards (both adults and juveniles) from the invaded population ate fire ants than did those from the uninvaded population. This suggests that exposure to fire ants over the past 37 generations (see 'Materials and methods' for calculation) has decreased population-level consumption of fire ants, at least while on a mound. The intrinsic decrease in this behaviour following invasion is reflected in the laboratory-reared lizards by the negative haldane estimates of per generation divergence in this trait (Fig. 3).

image

Figure 3. Ontogenetic shifts in lizard behaviour imposed by fire ants. In our study system, red imported fire ants impose greater bottom-up effects than top-down effects (relative strength of effects are indicated by thickness of arrows). Bottom-up pressure results from fire ant consumption by lizards, which results in envenomation through lizards being stung inside the mouth. Juveniles eat more ants than adults and are more vulnerable to the venom due to their smaller size. As a result, fire ant invasion may select for juveniles to avoid eating fire ants whereas adults, who are less affected by this mode of envenomation, may learn that it is safe for them to eat these abundant invaders. Top-down pressure results from fire ant predation on lizards. Top-down effects of fire ants on adults are relatively strong compared to effects on juveniles, because juveniles innately flee from fire ant attack. This flee behaviour is not retained into adulthood within uninvaded sites which consequently are at greater risk of mortality, but is exhibited by adults from invaded sites (along with twitching behaviour), possibly due to selection for this responsiveness (Langkilde, 2009, 2010). Estimates of phenotypic (hp) and intrinsic (hg, genetic and/or maternal) divergence in trait values are included where significant changes in traits were found between invaded and uninvaded populations (from Langkilde, 2009 and this study). NS denotes nonsignificant differences in trait values. Original artwork is by Taylor Olmsted, modified for this figure by T. R. Robbins.

Download figure to PowerPoint

Data from field-caught lizards revealed that this effect of evolutionary history with fire ants interacts with lifetime exposure to this invader to determine the propensity of lizards to eat fire ants. Field-caught lizards from the uninvaded population (having never experienced fire ants) exhibited a similar ontogenetic shift in fire ant consumption to that exhibited by laboratory-reared lizards (also having never experienced fire ants) from both populations; more juveniles ate fire ants than did adults. However, field-caught lizards from the fire ant-invaded population (that had been exposed to these ants both over many generations and within their lifetimes) exhibit the opposite pattern: more adults ate fire ants than did juveniles (Fig. 2). The ontogeny of fire ant-eating behaviour of lizards within fire ant-invaded sites opposes the general shift in diet away from fire ants with increasing age in this species.

The reversed ontogenetic shift likely arises from selection acting on juveniles to avoid eating fire ants, and a plastic response in adults that increases the consumption of fire ants. This apparent, opposing selection is reflected in the directions of phenotypic divergence in adults (positive) vs. juveniles (negative) following fire ant invasion, and is driven by the interaction between the effects of lifetime and evolutionary exposure to this novel threat (Figs 2 and 3). Maternal effects cannot be ruled out, but would have existed in both field-caught and laboratory-reared treatments, which minimizes the influence of maternal effects when comparing these treatments. Thus, the different invasion status × ontogenetic stage relationship between field-caught and laboratory-reared lizards (Fig. 2) was caused by the environment experienced by each treatment through ontogeny (fire ant exposure vs. no fire ant exposure).

Our results present two phenomena in need of explanation: (i) the intrinsic pattern of reduced fire ant consumption by lizards from the invaded population and (ii) the change of the ontogenetic shift in fire ant consumption by lizards due to lifetime exposure to fire ants. The intrinsic difference could be explained by selective pressure acting against fire ant consumption. Selection should be especially strong for juveniles, which have an ant-dominated diet and are more likely to succumb to the effects of venom due to their small size. The avoidance of fire ants by laboratory-reared adults from the invaded population suggests that the behaviour persists into adulthood in the absence of exposure to fire ants. But lifetime exposure to fire ants causes the reversal of this pattern.

Why do field-caught adults from the invaded population expose themselves to fire ant venom by eating fire ants? Ant-eating could be a learned antipredator strategy to survive fire ant attack by killing attacking ants (as per horned lizards; Webb & Henke, 2003), but eating fire ants during attack on the mound may lead to greater envenomation. Eating an ant scout away from the mound, however, would prevent attack from occurring, as the scout would then be unable to recruit additional ants from the colony to attack (Tschinkel, 2006; Freidenfelds et al., 2012). Alternatively (or concurrently), these adults may simply be learning to utilize fire ants as food. Adult fence lizards do eat ants, and their larger size means that they can likely eat many more fire ants than can juveniles and survive. Eating (as few as four) fire ants can cause delayed mortality in recent hatchling fence lizards (Langkilde & Freidenfelds, 2010). We do not know how many fire ants an adult lizard can consume without dying, and other longer-term costs of eating fire ants are unknown. For example, horned lizards (Phrynosoma coronatum) exhibit reduced growth rates when maintained on a diet of invasive Argentine ants (Linepithema humile) (or ants that are representative of an invaded community), as compared to a diet of native species (Suarez & Case, 2002); this is thought to be due to differences in catchability of ants.

Red imported fire ants out-compete native ant species, becoming the most abundant species within invaded sites (Porter & Savignano, 1990; but see King & Tschinkel, 2008), which makes them potentially a valuable food resource for adult lizards. Whether the observed ontogenetic shift in ant-eating behaviour following invasion is caused directly by the fire ants themselves or indirectly by the effect of fire ants on other potential prey, for example, is unknown. The mechanism responsible for this change may be selection for increased plasticity in ant-eating behaviour, which would allow adult lizards to take advantage of this new food resource. This plasticity may reflect ontogenetic conflict following fire ant invasion, whereby ant-eating behaviour is selected against in juveniles, but favoured in adults. Confirmation of increased plasticity and/or ontogenetic conflict as selective mechanisms, however, would require fitness values for juveniles and adults of varying ant-eating plasticity and/or propensity, which is beyond our current study.

We note that this study is of feeding behaviour while on a fire ant mound, and potentially under attack, which means feeding behaviour may be influenced by the context. Whether what we have measured is a true foraging behaviour or antipredator response cannot currently be determined. Examining foraging behaviour without the threat of attack (i.e. foraging behaviour in the presence of a single ant) could shed light on this issue (Robbins & Langkilde, in press). Nonetheless, the interaction between lifetime and evolutionary effects of exposure to fire ants suggests behavioural differences between these populations that are likely adaptive and are associated with consuming fire ants.

We should note that more laboratory-reared lizards consumed fire ants than field-caught lizards overall. We are unable to explain this pattern, but it is possible that lizards in the laboratory, which were reared on crickets, may have been more likely to eat fire ants during the behavioural trials simply because these ants represented a novel prey item (Desfilis et al., 2003). Alternatively, this pattern may be a sampling artefact caused by our laboratory-reared lizards having experienced no selective pressure vs. our field-caught lizards having already been subject to selection by fire ants (i.e. field-caught lizards that had a higher propensity to eat fire ants died as a result, prior to our sampling).

Integrating the results from this study with those from past research on this system offers a bigger picture of adaptive changes in multiple traits following invasion. This more comprehensive view suggests a complex ontogenetic shift in the selection imposed by fire ants, with stronger bottom-up effects of fire ants as venomous prey than top-down effects of fire ants as predators (Langkilde, 2009; see Fig. 3). Furthermore, the ontogenetic shift in selective pressures imposed by fire ants occurs in both top-down and bottom-up directions. The top-down pressure has resulted in adults behaving more like juveniles (Fig. 3), whereas the stronger bottom-up pressure has resulted in a reversal of the natural ontogenetic shift in relative behavioural roles through both evolutionary and plastic responses.

These data suggest that a single invader can impose very different selective pressures on a single native species, for example, having stronger bottom-up effects as venomous prey than top-down effects as predators (Fig. 3). We also provide evidence that learning through lifetime exposure to novel invasive threats can interact with evolutionary pressures to produce survival-enhancing behaviour that is specific to ontogenetic stage. Thus, this study contributes to our growing understanding of the complexities involved in behavioural responses (e.g. across ontogeny) to novel threats, how species are able to adapt and survive in the face of complex environmental change, and furthers our insights into the mechanisms of adaptation and coexistence.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

We thank B. Chitterling for comments on earlier drafts of the manuscript, and personnel at the Solon Dixon Forestry Education Center and St. Francis National Forest for logistical support. Financial support was provided by the National Science Foundation (DEB-0949483 to T. Langkilde). The research presented here adhered to Guidelines for the Use of Animals in Research and the Institutional Guidelines of Pennsylvania State University. Animal collection was authorized by the respective State's Permits.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information
  • Allen, C.R., Demarais, S. & Lutz, R.S. 1997. Effects of red imported fire ants on recruitment of whitetailed deer fawns. J. Wildl. Manage. 61: 911916.
  • Allen, C.R., Forys, E.A., Rice, K.G. & Wojcik, D.P. 2001. Effects of fire ants (Hymenoptera:Formicidae) on hatchling turtles and prevalence of fire ants on sea turtle nesting beaches in Florida. Fla. Entomol. 84: 250253.
  • Allen, C.R., Epperson, D.M. & Garmestani, A.S. 2004. Red imported fire ant impacts on wildlife: a decade of research. Am. Midl. Nat. 152: 88103.
  • Boronow, K.E. & Langkilde, T. 2010. Sublethal effects of invasive fire ant venom on a native lizard. J. Exp. Zool. 313A: 1723.
  • Bos, A.R., Gumanao, G.S., van Katwijk, M.M., Mueller, B., Saceda, M.M. & Tejada, R.L.P. 2011. Ontogenetic habitat shift, population growth, and burrowing behavior of the Indo-Pacific beach star, Archaster typicus (Echinodermata; Asteroidea). Mar. Biol. 158: 639648.
  • Brooks, J.L. & Dodson, S.I. 1965. Predation, body size, and composition of plankton. Science 150: 28.
  • Buhlman, K.A. & Coffman, G. 2001. Fire ant predation of turtle nests and implications for the strategy of delayed emergence. J. Elisha Mitchell Soc. 117: 94100.
  • Case, T.J. & Bolger, D.T. 1991. The role of introduced species in shaping the distribution and abundance of island reptiles. Evol. Ecol. 5: 272290.
  • Code of Federal Regulations 2012. Title 7. Part 301.81. Subpart - Imported Fire Ant. Available at: http://ecfr.gpoaccess.gov (accessed 15 July 2012).
  • Cooper, W. 2009. Flight initiation distance decreases during social activity in lizards (Sceloporus virgatus). Behav. Ecol. Sociobiol. 63: 17651771.
  • Crawley, M.J. 1992. Natural Enemies: The Population Biology of Predators, Parasites, and Diseases. Blackwell Scientific Publications, Oxford, MA.
  • Dabbert, C.B., Mitchell, R.B., Mueller, J.M., Forbes, A.R. & Treadway, J.H. 2002. Northern bobwhite and red imported fire ant interactions in the Texas coastal plains. Southwest. Entomol. 25: 105110.
  • Demarco, V.G. & Ferguson, G.W. 1985. Maximum prey size of an insectivorous lizard, Sceloporus undulatus garmani. Copeia 1985: 10771080.
  • Desfilis, E., Font, E. & Guillén-Salazar, F. 2003. Stimulus control of predatory behavior by the Iberian wall lizard (Podarcis hispanica, Sauria, Lacertidae): effects of familiarity with prey. J. Comp. Psychol. 117: 309316.
  • Ellner, S.P., Geber, M.A. & Hairston, N.G. 2011. Does rapid evolution matter? Measuring the rate of contemporary evolution and its impacts on ecological dynamics. Ecol. Lett. 14: 603614.
  • Epperson, D.M. & Heise, C.D. 2003. Nesting and hatchling ecology of gopher tortoises (Gopherus polyphemus) in southern Mississippi. J. Herpetol. 37: 315324.
  • Freidenfelds, N.A., Robbins, T.R. & Langkilde, T. 2012. Evading invaders: the effectiveness of a behavioral response acquired through lifetime exposure. Behav. Ecol. 23: 259664.
  • Gagliano, M., McCormick, M.I. & Meekan, M.G. 2007. Survival against the odds: ontogenetic changes in selective pressure mediate growth-mortality trade-offs in a marine fish. Proc. Royal Soc. B: Biol. Sci. 274: 15751582.
  • Garmestani, A.S., Allen, C.R., LaBram, J.A. & Peck, A.E. 2004. Laboratory observations of red imported fire ant (Hymenoptera: Formicidae) predation upon eastern tent caterpillars (Lepidoptera: Lasiocampidae). J. Entomol. Sci. 39: 472473.
  • Greenlees, M.J., Brown, G.P., Webb, J.K., Phillips, B.L. & Shine, R. 2006. Effects of an invasive anuran the cane toad (Bufo marinus) on the invertebrate fauna of a tropical Australian floodplain. Anim. Conserv. 9: 431438.
  • Greenlees, M.J., Phillips, B.L. & Shine, R. 2010. Adjusting to a toxic invader: native Australian frogs learn not to prey on cane toads. Behav. Ecol. 21: 966971.
  • Hairston, N.G. & Dillon, T.A. 1990. Fluctuating selection and response in a population of fresh-water copepods. Evolution 44: 17961805.
  • Hendry, A.P. & Kinnison, M.T. 1999. Perspective: the pace of modern life: measuring rates of contemporary microevolution. Evolution 53: 16371653.
  • Hoffmeister, T.S., Vet, L.E.M., Biere, A., Holsinger, K. & Filser, J. 2005. Ecological and evolutionary consequences of biological invasion and habitat fragmentation. Ecosystems 8: 657667.
  • King, J.R. & Tschinkel, W.R. 2008. Experimental evidence that human impacts drive fire ant invasions and ecological change. Proc. Nat. Acad. Sci. U. S. A. 105: 2033920343.
  • Krogh, M.G. & Schweitzer, S.H. 1999. Least terns nesting on natural and artificial habitats in Georgia, USA. Waterbirds 22: 290296.
  • Landers, J.L., Garner, J.A. & McRae, W.A. 1980. Reproduction of gopher tortoises (Gopherus polyphemus) in southwestern Georgia. Herpetologica 36: 353361.
  • Langerhans, R.B. 2007 Evolutionary consequences of predation: avoidance, escape, reproduction, and diversification. In: Predation in Organisms: A Distinct Phenomenon (A.M.T. Elewa, ed.), pp. 177220. Springer-Verlag, Germany.
  • Langkilde, T. 2009. Invasive fire ants alter behavior and morphology of native lizards. Ecology 90: 208217.
  • Langkilde, T. 2010. Repeated exposure and handling effects on the escape response of fence lizards to encounters with invasive fire ants. Anim. Behav. 79: 291298.
  • Langkilde, T. & Freidenfelds, N.A. 2010. Consequences of envenomation: red imported fire ants have delayed effects on survival but not growth of native fence lizards. Wildl. Res. 37: 566573.
  • Leache, A.D. & Reeder, T.W. 2002. Molecular systematics of the Eastern Fence Lizard (Sceloporus undulatus): a comparison of parsimony, likelihood, and Bayesian approaches. Syst. Biol. 51: 4468.
  • Lockley, T.C. 1995. Effect of imported fire ant predation on a population of the least tern – an endangered species. Southwest. Entomol. 20: 517519.
  • Losos, J.B., Schoener, T.W. & Spiller, D.A. 2004. Predator-induced behaviour shifts and natural selection in field-experimental lizard populations. Nature 432: 505508.
  • Malmquist, H.J., Snorrason, S.S., Skulason, S., Jonsson, B., Sandlund, O.T. & Jonasson, P.M. 1992. Diet differentiation in polymorphic arctic charr in Thingvallavatn, Iceland. J. Anim. Ecol. 61: 2135.
  • Mancinelli, G. 2010. Intraspecific, size-dependent variation in the movement behaviour of a brackish-water isopod: a resource-free laboratory experiment. Mar. Freshwat. Behav. Physiol. 43: 321337.
  • Menge, B.A. 1995. Joint bottom-up and top-down regulation of rocky intertidal algal beds in South-Africa. Trends Ecol. Evol. 10: 431432.
  • Menge, B.A., Daley, B.A. & Wheeler, P.A. (1996) Control of interaction strength in marine benthic communities. In: Food Webs: Integration of Pattern and Dynamics, (G.A. Polis & K.O. Winemiller, eds.), pp. 258274. Chapman and Hall, New York.
  • Miles, D.B., Noecker, R., Roosenburg, W.M. & White, M.M. 2002. Genetic relationships among populations of Sceloporus undulatus fail to support present subspecific designations. Herpetologica 58: 277292.
  • Mooney, H.A. & Cleland, E.E. 2001. The evolutionary impact of invasive species. Proc. Nat. Acad. Sci. 98: 54465451.
  • Moore, R.D., Griffiths, R.A., O'Brien, C.M., Murphy, A. & Jay, D. 2004. Induced defences in an endangered amphibian in response to an introduced snake predator. Oecologia 141: 139147.
  • Nelson, D.W.M., Crossland, M.R. & Shine, R. 2011. Foraging responses of predators to novel toxic prey: effects of predator learning and relative prey abundance. Oikos 120: 152158.
  • Parker, W.S. 1994. Demography of the fence lizard, Sceloporus undulatus, in northern Mississippi. Copeia 1994: 136152.
  • Porter, S.D. & Savignano, D.A. 1990. Invasion of polygene fire ants decimate native ants and disrupt arthropod community. Ecology 71: 20952106.
  • Puvanendran, V., Salies, K., Laurel, B. & Brown, J.A. 2004. Size-dependent foraging of larval Atlantic cod (Gadus morhua). Can. J. Zool. 82: 13801389.
  • Read, G.W., Lind, N.K. & Oda, C.S. 1978. Histamine release by fire ant (Solenopsis) venom. Toxicon 16: 361367.
  • Robbins, T.R & Langkilde, T. Native predator eats invasive toxic prey: evidence for increased incidence of consumption rather than aversion-learning. Biol. Inv., in press.
  • Santos Filho, P.D.S. 1997. The effect of body size on prey choice by Rivulus luelingi Seegers 1984 (Aplocheiloidei: Rivulidae). Rev. Bras. Biol. 57: 551562.
  • Smith, H.M., Applegarth, J.S. & Chiszar, D. 1992. Adaptive convergence in the lizard superspecies Sceloporus undulatus. Bull. Maryland Herpetol. Soc. 28: 123149.
  • Smith, J.E., Whelan, C.J., Taylor, S.J., Denight, M.L. & Stake, M.M. 2007. Novel predator-prey interactions: is resistance futile? Evol. Ecol. Res. 9: 433446.
  • Strauss, S.Y., Lau, J.A. & Carroll, S.P. 2006. Evolutionary responses of natives to introduced species: what do introductions tell us about natural communities? Ecol. Lett. 9: 357374.
  • Suarez, A.V. & Case, T.J. 2002. Bottom-up effects on persistence of a specialist predator: ant invasions and horned lizards. Ecol. Appl. 12: 291298.
  • Tschinkel, W.R. 2006. The Fire Ants. The Belknap Press of Harvard University Press, Cambridge.
  • Vinegar, M.B. 1975. Comparative aggression in Sceloporus virgatus, S. Undulatus consobrinus, and S. U. tristichus (Sauria: Iguanidae). Anim. Behav. 23: 279286.
  • Vitt, L.J. 2000. Ecological consequences of body size in neonatal and small-bodied lizards in the neotropics. Herpetol. Monogr. 14: 388400.
  • Wauters, L.A., Tosi, G. & Gurnell, J. 2002. Interspecific competition in tree squirrels: do introduced grey squirrels (Sciurus carolinensis) deplete tree seeds hoarded by red squirrels (S. vulgaris)? Behav. Ecol. Sociobiol. 51: 360367.
  • Webb, S.L. & Henke, S. 2003. Defensive strategies of Texas horned lizards (Phrynosoma cornutum) against red imported fire ants. Herpetol. Rev. 34: 327328.
  • Webb, J.K. & Shine, R. 1993. Prey-size selection, gape limitation and predator vulnerability in Australian blindsnakes (Typhlopidae). Anim. Behav. 45: 11171126.
  • Wetterer, J.K., Wood, L.D., Johnson, C., Krahe, H. & Fitchett, S. 2007. Predaceous ants, beach replenishment, and nest placement by sea turtles. Environ. Entomol. 36: 10841091.
  • Wilcoxen, T.E. & Rensel, M.A. 2009. Invasive fire ants depredate nest of Florida scrub-jay. Wilson J. Ornithol. 121: 846847.

Data deposited at Dryad: doi:10.5061/dryad.s3003

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  9. Supporting Information

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer-reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

FilenameFormatSizeDescription
jeb2583-sup-0001-MovieS1.mp4MPEG-4 video15637KMovie S1 Fence lizard eats three fire ants during encounter (movie credit: Tracy Langkilde, Katie Boronow and Travis R. Robbins).

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.