- Top of page
- Materials and methods
Life-history variation within a population is dependent on the type of the environment, namely a constant, stochastic or predictable environment (Roff, 2005). In a constant environment, the general expectation is that there would be a rapid erosion of phenotypic variation (Alpert & Simms, 2002). In a stochastic environment, no cues are available to indicate what the environment might be in the following generation, and the optimal value for a certain trait (i.e. selection itself) changes over time. Such fluctuating selection may favour the maintenance of phenotypic variance via phenotypic plasticity (Slatkin, 1974; Carroll et al., 2007), as well as genetic variance (McDonald & Ayala, 1974; Ellner & Hairston, 1994). In a predictable environment, the appropriate phenotype can be determined in advance, and the organism can develop this appropriate phenotype by the mechanism of phenotypic plasticity (Sultan & Stearns, 2005).
Phenotypic plasticity also affects the expression of stress-related responses. A lack of such plasticity may result in population extinction under stress (Gavrilets & Scheiner, 1993; Ancel, 1999). On the other hand, extensive phenotypic variability in organismal functions weakens the effects of directional selection imposed by stressful environments and thus lessens the opportunity for genetic assimilation and evolution of adaptations to stress (Fear & Price, 1998; Ancel, 2000; Huey et al., 2003).
In this study, we tested for differences in starvation endurance characterising pit-building antlions originating from different climatic regions: a Mediterranean climate, which is more benign and predictable, and a desert climate, which is harsher and more stochastic (Table 1). Starvation endurance is an important trait, especially in sit-and-wait predators, which experience fluctuations in prey arrivals much more than actively searching predators (Riechert, 1992). Sit-and-wait predators, like plants, are constrained in their ability to avoid stressful conditions (e.g. hunger and high temperatures) because of their limited mobility. However, this limited mobility brought forth the prediction that they should be able to cope better with stressful conditions than widely foraging predators (Huey & Pianka, 1981; Perry & Pianka, 1997). Even within the sit-and-wait predator group, one expects to find some variability in response to starvation. For instance, not all species and populations experience the same fluctuations in prey abundance. Similarly, some habitats are richer and/or more predictable than others (Rosenberg, 1987). This means that the response to starvation may also be dependent on population of origin. Indeed, Arnett and Gotelli (2003) have illustrated that northern antlion populations originating from temperate regions had better starvation endurance than southern populations of subtropical regions.
Table 1. Regional climatic data representative of the areas where antlions were collected
|Region/population||January temperature (ºC)||August temperature (ºC)||Annual precipitation (mm)||Annual precipitation CV (%)||Average humidity (%)|
When feeding is resumed after a period of starvation, a variety of species often exhibit growth compensation (Metcalfe & Monaghan, 2001); that is, an acceleration in growth to compensate for the lack of growth during the starvation period (e.g. Jespersen & Toft, 2003; Dmitriew & Rowe, 2005; Stoks et al., 2006). Growth compensation enables individuals to reach reproductive size despite experiencing a period of nutritional deficit, but may come at a cost that is apparent later in the individual's lifetime, such as decreased fecundity or life span (Metcalfe & Monaghan, 2001). In a prior study (Scharf et al., 2009), it was shown that antlions experiencing a longer period of starvation gained less mass after exploiting a newly provided prey than antlions experiencing a shorter starvation period. This pattern is incongruent with the phenomenon of growth compensation, and suggests the existence of some trade-off between fast growth and starvation endurance. We hypothesised that antlions from harsher environments, where the conditions are less stable and the chances of encountering prey are lower, will be less sensitive to starvation. More specifically, by using climate chambers to simulate different local conditions, we conducted a transplant experiment comparing responses to starvation in antlions (Myrmeleon hyalinus Olivier, 1811; Neuroptera: Myrmeleontidae) from Mediterranean and hyper-arid habitats. We expected loss of body mass during the starvation phase to be negatively correlated with the aridity gradient (highest in antlions from the Mediterranean climate and lowest in antlions from a hyper-arid climate). Additionally, if growth compensation does exist, then antlions that lose more mass during a starvation phase will also gain more mass when feeding is resumed (this prediction was, to some extent, refuted in a short-term experiment with M. hyalinus; see Scharf et al., 2009). We also predicted that desert populations, which should be more resistant to starvation, will exhibit less compensation in growth than Mediterranean populations, because of a longer rehabilitation period, requiring an increase of their metabolic rate.
Monitoring the antlions' growth rate during feeding and starvation phases enabled us to test for the effects of starvation on individuals from different populations. The effects of habitat (‘home’ versus ‘away’) enabled us to test whether the antlions' response to starvation is determined more by genetic effects or more by phenotypic plasticity. The effect of feeding (different feeding rates) enabled us to test whether higher prey encounter rates affect the response to a following starvation period, and whether they affect growth compensation when prey is encountered after a starvation period.
- Top of page
- Materials and methods
Our experiment tested for the effect of population of origin, current climatic conditions and feeding rate on starvation endurance in the pit-building antlion M. hyalinus. Contrary to our original hypothesis, population of origin had no significant effect on rates of body mass loss during starvation or on relative growth rates when feeding was resumed. It seems that despite the major differences in climatic conditions experienced by the two explored populations (Table 1), clear signals of genetically driven differences in starvation endurance between these populations could not be detected. Although one would expect organisms inhabiting harsher habitats, in terms of both biotic (less available prey) and abiotic (higher temperature, lower RH) factors, to be more resilient to prolonged starvation, this does not seem to be the case for M. hyalinus. Geographically based differences in starvation endurance between antlion populations were previously described by Arnett and Gotelli (2003), and are expected to arise due to body mass differences (which were shown to exist between M. hyalinus populations in Israel; Scharf et al., 2008), and also due to latitudinal differences, even after factoring out initial body size (Arnett & Gotelli, 2003). In another study, antlions from an arid and less productive climate were found to lose less mass during starvation, and recover faster after the starvation period, compared with antlions originating from a semi-arid climate (Rotkopf et al., 2013).
Food availability often limits body size and growth in ectotherms (Reznick, 1990; Niewiarowski, 1995). For example, in controlled laboratory experiments, growth rates and size at maturity of guppies (Reznick, 1990) and antlions (Arnett & Gotelli, 1999b) increased at higher food levels. In eastern North America, larval and adult body size of the antlion Myrmeleon immaculatus De Geer, 1773, increases with latitude (Arnett & Gotelli, 1999a). In a common-garden laboratory experiment, antlion larvae were collected from southern (Georgia, South Carolina) and northern (Connecticut, Rhode Island) populations and reared under different environmental conditions. Larvae reared with increased food levels grew faster and achieved a larger body size, regardless of temperature or population source (Arnett & Gotelli, 1999b). In a controlled field experiment, larvae in an Oklahoma population also grew faster and reached a larger adult body size with food supplements (Gotelli, 1996). These studies established that food availability can influence larval and adult body size in antlions. Indeed, in our experiment, the most significant factor was feeding, which, apart from the obvious effect of faster growth rate in the initial feeding phase, also affected mass loss rate during starvation, and relative growth rate during the compensation phase. This is a classic example of growth compensation (Metcalfe & Monaghan, 2001), as antlions that were fed less frequently during the initial feeding phase compensated best when food supply was renewed after starvation. A similar effect was found in a recent study with another antlion species, Cueta lineosa (Rotkopf et al., 2013).
Notably, regarding the rate of mass loss during starvation, our results differ from those found in a previous study on this same species (Scharf et al., 2009), considering only a single semi-arid population. Specifically, Scharf et al. (2009) illustrated that antlions growing faster during the feeding phase also lost mass faster during the successive starvation phase, and presented higher growth efficiency during the compensation phase. These results, which are incongruent with the growth compensation phenomenon, led the authors to suggest the existence of an induced trade-off between fast growth and starvation endurance. A possible explanation for the discrepancy of results between the two experiments is the large difference in the duration of the starvation phases: Scharf et al. (2009) starved the antlions for a 2-week period, while in our experiment the antlions were starved for 2 months. We believe that since antlions are adapted to long starvation periods, a 2-month starvation period is more representative of their physiological responses to starvation stress. Considering the effects of growth rate, fast growth is generally thought to decrease an organism's resistance to shortage in prey supply (i.e. low starvation endurance), as was illustrated in butterflies (Stockhoff, 1991; Gotthard et al., 1994). A probable explanation is the induction of metabolic processes that demand energy and will deplete the stored reserves faster in a period of starvation (Gotthard, 2001). Indeed, Stoks et al. (2006) have tested this explanation in a damselfly system, illustrating that fast-growing damselflies had a higher metabolic rate that was also associated with a faster depletion of their stored reserves (i.e. glycogen and triglycerides). In addition, Fischer et al. (2005) have shown that fast-growing butterflies lose more body mass during metamorphosis, possibly owing to higher respiration rate, supporting the idea that increased metabolic rate can be associated with a faster depletion of storage molecules. Here we also found that increased growth rate decreased starvation endurance, but this is mostly due to decreased compensation abilities and not because of increased mass loss during starvation.
The experimental conditions in the different climate treatments seemed to exert a significant effect on growth rate only during the starvation phase, in which antlions in the desert climate treatment, as expected, lost mass faster than those in the Mediterranean climate treatment. This difference in mass loss is probably a result of increased water loss, as well as increased resource utilization. The climate treatment was also found to affect mortality and pupation rates. Antlions experiencing desert conditions were expected to show higher mortality rates, but, interestingly, they also showed higher pupation rates, indicating that antlions under stress have the ability to ‘make the best out of a bad situation’ and pupate at earlier stages than usual. In a previous study (with regular feeding), southern desert populations of M. hyalinus pupated faster (and at smaller body masses) than northern Mediterranean populations, and larvae in the desert climate treatment pupated faster (and at smaller body masses) than larvae in the Mediterranean climate treatment (Scharf et al., 2008). It is possible that the starvation stress in our experiments increased pupation rates in the Mediterranean population, masking the differences in pupation rate between populations. A pattern of higher plasticity in the Mediterranean population, as opposed to a fixed response in the desert population, is consistent with previous studies, showing that populations accustomed to stressful conditions show little change in life-history traits, even when exposed to better conditions (Ward & Slotow, 1992; Niewiarowski & Roosenburg, 1993).
In summary, our study found little evidence for differences between populations of M. hyalinus in their response to starvation. It is possible that this species' natural microhabitat (shaded areas under trees or large bushes) acts as a buffer from the extreme conditions of the desert environment, and that the difference in response to starvation is apparent not in the antlions' mass loss rate, but in increased plasticity in life-history decisions, such as pupation time, in the Mediterranean population.
This study emphasises the importance of testing responses to stress when comparing life-history decisions in individuals from different populations. Some phenotypic differences between populations might be apparent only when exposing the experimental organisms to external stress (in addition to the environmental change the organism experiences in a common-garden or transplant experiment), while, on the other hand, phenotypic differences apparent under stress-free conditions might be masked by the effects of the stress factor, which also leads to important findings regarding the level of plasticity in response to stress in different populations.
Future directions for this research should include an exploration of the physiological and molecular mechanisms underlying M. hyalinus's tolerance to starvation, and differences in these mechanisms between populations originating from different climatic regions. These mechanisms could include cuticle lipid composition and permeability to water loss (Gibbs, 1998, 2002), differences in metabolic rates and metabolic fuel utilization (e.g. Kalra & Gefen, 2012), and different expression levels of genes involved in distinct physiological pathways mediating sugar and fat metabolism, and cell growth (Zinke et al., 2002).