Marjan De Block

Life history plasticity in a damselfly : Effects of combined time and biotic constraints

Optimal values for life history traits are expected to depend upon environ- mental conditions during development and the period within which development is con- strained (e.g., biotic factors and time constraints, respectively). Theory predicts that life history responses to both biotic factors and time constraints may be both direct and be- haviorally mediated. Few experimental studies of life histories have considered the joint effects of biotic factors and time constraints, and fewer still have been able to disentangle direct from behaviorally mediated effects. We studied such interactions by manipulating the perceived time to the onset of winter, predation risk, and food resources level in larvae of the damselfly Lestes sponsa. In the first experiment (predation 3 time constraint), the presence of a predator caused an overall reduction in foraging activity, development rate, and mass at emergence. However, larvae that had less time available before the end of the season, increased foraging activity and development rate, while mass at emergence de- creased. These results suggest that the observed changes in life history characters were behaviorally mediated in the presence of predators. In contrast, life history responses of time-constrained larvae occurred independently of the behavioral changes and, therefore, were direct. In the second experiment (food level 3 time constraint) larvae under high food levels had a higher foraging activity, increased development rate, and higher growth rates, compared to low food-level treatments. Time-constrained larvae accelerated devel- opment and had a smaller mass at emergence at high food levels than larvae that were not time constrained. In contrast, and opposite to predictions, time-constrained larvae at low food levels had the slowest development rate and the largest mass at emergence. We suggest that larvae in the latter group were aiming to delay emergence to the next season (cohort splitting). Our results suggest that both behaviorally mediated and direct responses to biotic factors and time constraints are a feature of the life history of this damselfly.

FITNESS EFFECTS FROM EGG TO REPRODUCTION: BRIDGING THE LIFE HISTORY TRANSITION

Although complex life cycles are widespread, we know little about how constraints in the larval stage influence adult fitness. Most models assume a tight coupling of larval conditions and adult fitness through size and timing of the life history transition. However, there are few empirical tests of this assumption. We combined an experimental manipulation of larval environment with a subsequent study of adult fitness, measured as lifetime mating success. Individuals of the damselfly Lestes viridis were followed from the egg stage to adult reproduction and death. Under time constraints, emergence occurred earlier, but in late-hatched larvae, this did not result in a lower size. Under nutritional constraints, emergence occurred later, and size was reduced. Variation in survival to maturity was better explained by larval constraints than by emergence traits, whereas both larval constraints and emergence traits explained variation in lifetime mating success. Sexes reacted differently to larval constraints, and the coupling of larval constraints to adult fitness also was sex specific. Our results indicate that larval constraints do not necessarily carry over to adult fitness through size and timing of transition, and that carryover effects may be sex specific. This may explain the existence of hidden costs that become visible after maturation and may explain part of the unexplained variation in selection studies on adults.

Compensatory growth and oxidative stress in a damselfly.

Physiological costs of compensatory growth are poorly understood, yet may be the key components in explaining why growth rates are typically submaximal. Here we tested the hypothesized direct costs of compensatory growth in terms of oxidative stress. We assessed oxidative stress in a study where we generated compensatory growth in body mass by exposing larvae of the damselfly Lestes viridis to a transient starvation period followed by ad libitum food. Compensatory growth in the larval stage was associated with higher oxidative stress (as measured by induction of superoxide dismutase and catalase) in the adult stage. Our results challenge two traditional views of life-history theory. First, they indicate that age and mass at metamorphosis not necessarily completely translate larval stress into adult fitness and that the observed physiological cost may explain hidden carry-over effects. Second, they support the notion that costs of compensatory growth may be associated with free-radical-mediated trade-offs and not necessarily with resource-mediated trade-offs.

Life-history variation in relation to time constraints in a damselfly

Although variation within populations in plasticity to time constraints is expected with regard to hatching date, empirical studies are largely lacking. We studied life-history responses to time constraints manipulated by photoperiod and associated with hatching date in larvae of the damselfly Lestes viridis for two populations with a different hydroperiod. In a common garden experiment, early- and late-hatched larvae from both populations were reared at two photoperiods mimicking the start and the end of the egg-hatching season. In a reciprocal transplant experiment, early- and late-hatched larvae from both populations were reared in both ponds. In all these experiments, larvae were reared from egg hatching until adult emergence. Within both populations, larvae reared at the photoperiod indicating a late time point in the growing season, reduced development time to compensate for their perceived shorter development period. Growth rate, however, did not respond to photoperiod, resulting in a lower mass at emergence. As expected, both in the laboratory and in the field, larvae from eggs that hatched later in the season generally had a shorter development time and a faster growth rate, resulting in a higher mass at emergence compared to early-hatched larvae. This may explain the intriguing seasonal increase in mass at emergence in this species, and affect the predictions of optimality models. None of these life-history responses differed between the two populations, despite clear differences in time constraints linked to hydroperiod, suggesting the robustness of the observed patterns. Given the ubiquity of asynchronous hatching in nature, and the adaptive value of the observed differences between early- and late-hatched larvae, we expect the effects of hatching date on life-history plasticity to be widespread.

Lethal and sublethal costs of autotomy and predator presence in damselfly larvae

Abstract We studied the costs of lamellae autotomy with respect to growth and survival of Lestes sponsa damselfly larvae in field experiments. We manipulated predation risk by Aeshna cyanea dragonfly larvae and lamellae status of L. sponsa larvae in field enclosures and compared differences in numbers, size and mass of survivors among treatments. In the absence of a free-ranging A. cyanea larva, about 29% of the L. sponsa larvae died. This was probably due to cannibalism. The presence of a free-ranging A. cyanea reduced larval survival by 68% compared to treatments in which it was absent or not permitted to forage on L. sponsa damselflies. Across all predator treatments, lamellae autotomy reduced survival by about 20%. The mean head width and mass of survivors was lower in the enclosures with a free-ranging A. cyanea compared to the other two predator treatments. This suggested that larvae grew less in the presence of a free-ranging predator, indicating that increased antipredator behaviours were more important in shaping growth responses than reduced population density. Mass, but not head width, of survivors was also reduced after autotomy. The fitness consequences of these effects for the adults may be pronounced. In general, these field data strongly suggest that lamellae autotomy affects population regulation of damselflies.

Genetic differentiation and dispersal among populations of the damselfly **Lestes viridis** (Odonata)

We investigated genetic differentiation among 8 populations of the protected damselfly Lestes viridis (Vander Linden, 1825) in permanent ponds in northern Belgium by means of allozyme electrophoresis and isoelectric focusing, and estimated levels of gene flow using F-statistics. In addition, we did a capture-mark-recapture experiment to estimate direct levels of gene flow. Our aim was to test whether populations of L. viridis represented a single, large panmictic population or formed a series of demographically isolated populations, which may be defined as appropriate management units (MU). None of the marked individuals moved among the ponds, indicating a strong fidelity for adults to their breeding pond. Only 1 population was genetically strongly differentiated, whereas little or no differentiation was observed among the 7 other ponds. Absence of adult dispersal but genetic homogeneity between ponds thus suggested substantial teneral dispersal. We observed large heterozygote deficiencies at 2 loci (Fdh and Est). Significant differences in allele frequencies among ponds suggests that populations of L. viridis in northern Belgium may consist of >1 MU. Management plans for conservation should therefore take into consideration the presence of several MUs in this species.

Autotomy reduces immune function and antioxidant defence

Costs of autotomy, an antipredator defence, are typically explained by impaired mobility; yet physiologically mediated costs may also play a role. Given the resemblance to wounding, a decreased immune function and an associated reduction in antioxidant defence is expected after autotomy. In line with this, after lamellae autotomy, larvae of the damselfly Lestes viridis showed lower levels of innate immunity (i.e. phenoloxidase, PO) and antioxidant defence (superoxide dismutase, SOD). Levels of catalase (CAT) remained, however, unaffected. In line with its cytotoxicity, PO covaried positively with CAT, yet negatively with SOD. We identified a novel cost of autotomy in terms of a reduced innate immunity, which may provide an alternative explanation for the often observed costs of autotomy and which may generate indirect interactions between predators and parasites.

Spatiotemporal allozyme variation in the damselfly, Lestes viridis (Odonata: Zygoptera): gene flow among permanent and temporary ponds

Several insect species seem to persist not only in permanent but also in temporary ponds where they face particularly harsh conditions and frequent extinctions. Under such conditions, gene flow may prevent local adaptation to temporary ponds and may promote phenotypic plasticity, or maintain apparent population persistence. The few empirical studies on insects suggest the latter mechanism, but no studies so far quantified gene flow including both pond types. We investigated the effects of pond type and temporal variation on population genetic differentiation and gene flow in the damselfly Lestes viridis in northern Belgium. We report a survey of two allozyme loci (Gpi, Pgm) with polyacrylamide gel electrophoresis in 14 populations from permanent and temporary ponds, and compared these results with similar data from the same permanent populations one year before. The data suggested that neither pond-drying regime, nor temporal variation have a substantial effect on population genetic structuring and did not provide evidence for stable population differentiation in L. viridis in northern Belgium. Gene flow estimates were high within permanent and temporary ponds, and between pond types. Our data are consistent with a source-sink metapopulation system where temporary ponds act as sinks in dry years, and are quickly recolonized after local population extinction. This may create a pattern of apparent population persistence of this species in permanent and temporary ponds without clear local adaptation.

Phenotypic shifts caused by predation: selection or life-history shifts?

Predators can impose both selection and life-history shifts in prey populations. Because both processes may affect phenotypic distributions, the estimates of selection differentials may be biased. We carried out two field experiments to disentangle these separate effects. We studied whether dragonfly predation by Aeshna cyanea changes the distributions in body size and lamellae morphology in the damselfly Lestes sponsa. Damselflies have caudal lamellae which are used in escapes by swimming. In a first experiment, we manipulated predator presence (No Aeshna, Encaged Aeshna or Free-ranging Aeshna) and stopped the experiment when all larvae had moulted once. In a second experiment, larvae were confronted with a Free-ranging Aeshna but collected before moulting, and survivors were compared with a control sample taken at the start of the experiment. The presence of Aeshna largely reduced the survival probabilities of the Lestes larvae at a very similar rate in both experiments. Daily survival probabilities did not differ between the No Aeshna and Encaged Aeshna treatments. In the Free-ranging Aeshna treatment of the first experiment, size was reduced compared to the other two treatments, creating a significant apparent selection differential. This was probably mainly due to predator-induced reduced growth because in the second experiment, where growth effects were excluded, size of the survivors did not differ from the control sample. In both experiments there was a significant selection pressure for larger lamellae. Standardized directional selection differentials were similar in both experiments (0.57 and 0.28 phenotypic standard deviation units). No survival selection on lamellae shape was detected. These results are in agreement with previous findings that lamellae size, but not lamellae shape, enhances swimming performance and thereby predator escape in this species.

Egg hatching patterns within and among populations of a damselfly occupying permanent and temporary ponds

Although opposing selection forces cause drastic differences in community structure between temporary and permanent ponds, some species are able to persist in both pond types. Little is known about the underlying life history strategies that enable species to do so. This is especially true for embryonic development times. Here, we describe within and among population variation in natural egg hatching patterns of the damselfly Lestes viridis that occurs in both pond types. In general, egg hatching was synchronous both within and among populations. A two-year field monitoring study showed consistent regional differences in egg hatching and earlier egg hatching in temporary ponds. A common garden and two full-sib experiments suggested that differences in hatching dates among populations and families were not completely driven by differences in environmental conditions, but may have a genetic basis. Although the pattern of earlier egg hatching in temporary ponds, as observed in the field monitoring, is adaptive, it was not fully repeatable in the common garden experiment. This suggests that this pattern is caused by more benign environmental conditions at the temporary ponds relative to the permanent ponds, and not an adaptation to pond type.