Both strong directional selection and faster development are thought to destabilize development, giving rise to greater fluctuating asymmetry (FA), although there is no strong empirical evidence supporting this assertion. We compared FA in sternopleural bristle number in four populations ofDrosophila melanogaster successfully selected for faster development from egg to adult, and in four control populations. The fraction of perfectly symmetric individuals was higher in the selected populations, whereas the FA levels did not differ significantly between selected and control populations, clearly indicating that directional selection for faster development has not led to increased FA in sternopleural bristle number in these populations. This may be because: (i) development time and FA are uncorrelated, (ii) faster development does result in FA, but selection has favoured developmentally stable individuals that can develop fast and still be symmetrical, or (iii) the increased fraction of symmetric individuals in the selected populations is an artifact of reduced body size. Although we cannot discriminate among these explanations, our results suggest that the relationship between development time, FA and fitness may be far more subtle than often thought.

In theDrosophila literature, selection for faster development and selection for adapting to high density are often confounded, leading, for example, to the expectation that selection for faster development should also lead to higher competitive ability. At the same time, results from experimental studies on evolution at high density do not agree with many of the predictions from classical density-dependent selection theory. We put together a number of theoretical and empirical results from the literature, and some new experimental results onDrosophila populations successfully subjected to selection for faster development, to argue for a broader interpretation of density-dependent selection. We show that incorporating notions of α-selection, and the division of competitive ability into effectiveness and tolerance components, into the concept of density-dependent selection yields a formulation that allows for a better understanding of the empirical results. We also use this broader formulation to predict that selection for faster development inDrosophila should, in fact, lead to the correlated evolution of decreased competitive ability, even though it does lead to the evolution of greater efficiency and higher population growth rates at high density when in monotypic culture.

A series of laboratory selection experiments onDrosophila melanogaster over the past two decades has provided insights into the specifics of life-history tradeoffs in the species and greatly refined our understanding of how ecology and genetics interact in life-history evolution. Much of what has been learnt from these studies about the subtlety of the microevolutionary process also has significant implications for experimental design and inference in organismal biology beyond life-history evolution, as well as for studies of evolution in the wild. Here we review work on the ecology and evolution of life-histories in laboratory populations ofD. melanogaster, emphasizing how environmental effects on life-history-related traits can influence evolutionary change. We discuss life-history tradeoffs—many unexpected—revealed by selection experiments, and also highlight recent work that underscores the importance to life-history evolution of cross-generation and cross-life-stage effects and interactions, sexual antagonism and sexual dimorphism, population dynamics, and the possible role of biological clocks in timing life-history events. Finally, we discuss some of the limitations of typical selection experiments, and how these limitations might be transcended in the future by a combination of more elaborate and realistic selection experiments, developmental evolutionary biology, and the emerging discipline of phenomics.

The relative contributions of ancestry, chance, and past and ongoing election to variation in one adaptive (larval feeding rate) and one seemingly nonadaptive (pupation height) trait were determined in populations ofDrosophila melanogaster adapting to either low or high larval densities in the laboratory. Larval feeding rates increased rapidly in response to high density, and the effects of ancestry, past selection and chance were ameliorated by ongoing selection within 15–20 generations. Similarly, in populations previously kept at high larval density, and then switched to low larval density, the decline of larval feeding rate to ancestral levels was rapid (15-20 generations) and complete, providing support for a previously stated hypothesis regarding the costs of faster feeding inDrosophila larvae. Variation among individuals was the major contributor to variation in pupation height, a trait that would superficially appear to be nonadaptive in the environmental context of the populations used in this study because it did not diverge between sets of populations kept at low versus high larval density for many generations. However, the degree of divergence among populations (FST) for pupation height was significantly less than expected for a selectively neutral trait, and we integrate results from previous studies to suggest that the variation for pupation height among populations is constrained by stabilizing selection, with a flat, plateau-like fitness function that, consequently, allows for substantial phenotypic variation within populations. Our results support the view that the genetic imprints of history (ancestry and past selection) in outbreeding sexual populations are typically likely to be transient in the face of ongoing selection and recombination. The results also illustrate the heuristic point that different forms of selection-for example directional versus stabilizing selection—acting on a trait in different populations may often not be due to differently shaped fitness functions, but rather due to differences in how the fitness function maps onto the actual distribution of phenotypes in a given population. We discuss these results in the light of previous work on reverse evolution, and the role of ancestry, chance, and past and ongoing selection in adaptive evolution.

Dry weight at eclosion, adult lifespan, lifetime fecundity, lipid and carbohydrate content at eclosion, and starvation and desiccation resistance at eclosion were assayed on a long-term laboratory population ofDrosophila melanogaster, and one recently wild-caught population each of four other species ofDrosophila, two from themelanogaster and two from theimmigrans species group. The relationships among trait means across the five species did not conform to expectations based on correlations among these traits inferred from selection studies onD. melanogaster. In particular, the expected positive relationships between fecundity and size/lipid content, lipid content and starvation resistance, carbohydrate (glycogen) content and desiccation resistance, and the expected negative relationship between lifespan and fecundity were not observed. Most traits were strongly positively correlated between sexes across species, except for fractional lipid content and starvation resistance per microgram lipid. For most traits, there was evidence for significant sexual dimorphism but the degree of dimorphism did not vary across species except in the case of adult lifespan, starvation resistance per microgram lipid, and desiccation resistance per microgram carbohydrate. Overall,D. nasuta nasuta andD. sulfurigaster neonasuta (immigrans group) were heavier at eclosion than themelanogaster group species, and tended to have somewhat higher absolute lipid content and starvation resistance. Yet, these twoimmigrans group species were shorter-lived and had lower average daily fecundity than themelanogaster group species. The smallest species,D. malerkotliana (melanogaster group), had relatively high daily fecundity, intermediate lifespan and high fractional lipid content, especially in females.D. ananassae (melanogaster group) had the highest absolute and fractional carbohydrate content, but its desiccation resistance per microgram carbohydrate was the lowest among the five species. In terms of overall performance, the laboratory population ofD. melanogaster was clearly superior, under laboratory conditions, to the other four species if adult lifespan, lifetime fecundity, average daily fecundity, and absolute starvation and desiccation resistance are considered. This finding is contrary to several recent reports of substantially higher adult lifespan and stress resistance in recently wild-caught flies, relative to flies maintained for a long time in discretegeneration laboratory cultures. Possible explanations for these apparent anomalies are discussed in the context of the differing selection pressures likely to be experienced byDrosophila populations in laboratory versus wild environments.

InD. melanogaster, the observation of greater pupation height under constant darkness than under constant light has been explained by the hypothesis that light has an inhibitory effect on larval wandering behaviour, preventing larvae from crawling higher up the walls of culture vials prior to pupation. If this is the only role of light in affecting pupation height, then various light : dark regimes would be predicted to yield pupation heights intermediate between those seen in constant light and constant darkness. We tested this hypothesis by measuring pupation height under various light : dark regimes in four laboratory populations ofDrosophila melanogaster. Pupation height was the greatest in constant darkness, intermediate in constant light, and the least in a light/dark regime of LD 14:14 h. The results clearly suggest that there is more to light regime effects on pupation height than mere behavioural inhibition of wandering larvae, and that circadian organization may play some role in determining pupation height, although the details of this role are not yet clear. We briefly discuss these results in the context of the possible involvement of circadian clocks in life-history evolution.

Models of competitor coevolution, especially the genetic feedback hypothesis, suggest that a negative correlation between intraspecific and interspecific competitive effects may be important in sustaining competitor coexistence, and can give rise to oscillatory dynamics with repeated reversals of competitive superiority. I reanalyzed previously published census data from an experiment in which populationsof Drosophila melanogaster andD. simulans underwent competitive coevolution in one familiar and two novel environments, to specifically look for any evidence of a negative relationship between intraspecific and interspecific competitive effects on population growth rates, and for any indication of short period cycling in the relative magnitude of intraspecific and interspecific competitive effects. While there was considerable variation in the relative magnitude of intraspecific and interspecific competitive effects over generations, among both populations and environments, there was no clear evidence supporting the genetic feedback hypothesis. Intraspecific and interspecific competitive effects on population growth rates were strongly positively correlated in novel environments, and uncorrelated in the familiar environment. Data from the familiar environment indicated that indices of competition of populations of the initially superior competitor,D. melanogaster, might be showing some cyclic behaviour, but I argue that this is likely to be transient, and not suggestive of sustained oscillatory dynamics predicted by the genetic feedback model. I discuss the results in the context of the importance of the genetic architecture of intraspecific and interspecific competitive abilities in determining the coevolutionary trajectory of competitive interactions.

Census population size, sex-ratio and female reproductive success were monitored in 10 laboratory populations of Drosophila melanogaster selected for different ages of reproduction. With this demographic information, we estimated eigenvalue, variance and probability of allele loss effective population sizes. We conclude that estimates of effective size based on genefrequency change at a few loci are biased downwards. We analysed the relative roles of selection and genetic drift in maintaining genetic variation in laboratory populations of Drosophila. We suggest that rare, favourable genetic variants in our laboratory populations have a high chance of being lost if their fitness effect is weak, e.g. 1% or less. However, if the fitness effect of this variation is 10% or greater, these rare variants are likely to increase to high frequency. The demographic information developed in this study suggests that some of our laboratory populations harbour more genetic variation than expected. One explanation for this finding is that part of the genetic variation in these outbred laboratory Drosophila populations may be maintained by some form of balancing selection. We suggest that, unlike bacteria, medium-term adaptation of laboratory populations of fruit flies is not primarily driven by new mutations, but rather by changes in the frequency of preexisting alleles.

The standard view of adaptation to larval crowding in fruitflies, built on results from 25 years of multiple experimental evo-lution studies onDrosophila melanogaster

, was that enhanced competitive ability evolves primarily through increased larvalfeeding and foraging rate, and increased larval tolerance to nitrogenous wastes, at the cost of efficiency of food conversion tobiomass. These results were at odds from the predictions of classicalK

-selection theory, notably the expectation that selec-tion at high density should result in the increase of efficiency of conversion of food to biomass, and were better interpretedthrough the lens of

α

-selection. We show here that populations ofD. ananassaeandD. n. nasutasubjected to extreme larvalcrowding evolve greater competitive ability and pre-adult survivorship at high density, primarily through a combination ofreduced larval duration, faster attainment of minimum critical size for pupation, greater time efficiency of food conversion tobiomass and increased pupation height, with a relatively small role of increased urea/ammonia tolerance, if at all. This is avery different suite of traits than that seen to evolve under similar selection inD

.melanogaster

,andseemstobeclosertotheexpectations from the canonical theory ofK

-selection. We also discuss possible reasons for these differences in results acrossthe three species. Overall, the results reinforce the view that our understanding of the evolution of competitive ability in fruit-flies needs to be more nuanced than before, with an appreciation that there may be multiple evolutionary routes through whichhigher competitive ability can be attained.

Multiple experimental evolution studies on Drosophila melanogasterin the 1980s and 1990s indicated that enhanced competitive ability evolved primarily through increased larval tolerance to nitrogenous wastes and increased larval feeding and foraging rate, at the cost of efficiency of food conversion to biomass, and this became the widely accepted view of how adaptation to larval crowding evolves in fruitflies. We recently showed that populations of D. ananassaeand D. n. nasuta subjected to extreme larval crowding evolved greater competitive ability without evolving higher feeding rates, primarily through acombination of reduced larval duration, faster attainment of minimum critical size for pupation, greater efficiency of food conversion to biomass, increased pupation height and, perhaps, greater urea/ammonia tolerance. This was a very differentsuite of traits than that seen to evolve under similar selection in D. melanogasterand was closer to the expectations from the theory of K-selection. At that time, we suggested two possible reasons for the differences in the phenotypic correlates ofgreater competitive ability seen in the studies with D. melanogaster and the other two species. First, that D. ananassae and D. n. nasuta had a very different genetic architecture of traits affecting competitive ability compared to the long-term labora-tory populations of D. melanogaster used in the earlier studies, either because the populations of the former two species were relatively recently wild-caught, or by virtue of being different species. Second, that the different evolutionary trajectories in D. ananassae and D. n. nasuta versus D. melanogaster were a reflection of differences in the manner in which larval crowding was imposed in the two sets of selection experiments. The D. melanogaster studies used a higher absolute density of eggs per unit volume of food, and a substantially larger total volume of food, than the studies on D. ananassae and D. n. nasuta. Here, we show that long-term laboratory populations of D. melanogaster , descended from some of the populations used in the earlier studies, evolve essentially the same set of traits as the D. ananassae and D. n. nasuta crowding-adapted populations whensubjected to a similar larval density at low absolute volumes of food. As in the case of D. ananassae and D. n. nasuta ,andin stark contrast to earlier studies with D. melanogaster , these crowding-adapted populations of D. melanogaster did not evolve greater larval feeding rates as a correlate of increased competitive ability. The present results clearly suggest that the suite of phenotypes through which the evolution of greater competitive ability is achieved in fruitflies depends critically not just on larval density per unit volume of food, but also on the total amount of food available in the culture vials. We discuss these results in the context of an hypothesis about how larval density and the height of the food column in culture vials might interact to alter the fitness costs and benefits of increased larval feeding rates, thus resulting in different routes to the evolution of greater competitive ability, depending on the details of exactly how the larval crowding was implemented.

Click here to see the Video abstract