Long-term selection for a quantitative character in large replicate populations of Drosophila melanogaster : 1. Response to selection

Thinking About the Evolution of Complex Traits in the Era of Genome-Wide Association Studies

Many traits of interest are highly heritable and genetically complex, meaning that much of the variation they exhibit arises from differences at numerous loci in the genome. Complex traits and their evolution have been studied for more than a century, but only in the last decade have genome-wide association studies (GWASs) in humans begun to reveal their genetic basis. Here, we bring these threads of research together to ask how findings from GWASs can further our understanding of the processes that give rise to heritable variation in complex traits and of the genetic basis of complex trait evolution in response to changing selection pressures (i.e., of polygenic adaptation). Conversely, we ask how evolutionary thinking helps us to interpret findings from GWASs and informs related efforts of practical importance.

Responses to relaxed and reverse selection in strains artificially selected for duration of death-feigning behavior in the red flour beetle, Tribolium castaneum

Divergent lines selected artificially for many generations make it possible to answer two questions: (1) whether genetic variation still exists within the selected population; and (2) whether the selection itself is costly for the selected strain. In previous studies, the red flour beetle Tribolium castaneum was divergently selected artificially for duration of death-feigning, and strains selected for longer (L-strain) and shorter (S-strain) durations of death-feigning have been established (Miyatake et al. 2004, 2008). Because the selection experiments have been conducted for more than 27 generations, genetic variation may be eroded. Furthermore, because another previous study reported physiological costs to L-strains, the L-strains selected artificially for longer duration of death-feigning may have suffered more costs than the S-strains. In the present study, therefore, we relaxed the selection pressure after the 27th or 30th generation of S- and L-strains. We also carried out reverse selection during the most recent eight generations of S- and L-strains. The results showed that each strain clearly responded to relaxation of selection and reverse selection, suggesting that (1) additive genetic variation still existed in both strains after long-term selection, and (2) selection for shorter and longer duration of death-feigning was costly. These results suggest that anti-predator behavior is controlled by many loci, and longer or shorter duration of death-feigning is costly in a laboratory without predators.

Symposium review: Breeding a better cow-Will she be adaptable?

Adaption is a process that makes an individual or population more suited to their environment. Long-term adaptation is predicated on ample usable genetic variation. Evolutionary forces influencing the extent and dynamics of genetic variation in a population include random drift, mutation, recombination, selection, and migration; the relative importance of each differs by population (i.e., drift is likely to be more influential in smaller populations) and number of generations exposed to selection (i.e., mutation is expected to contribute substantially to genetic variability following many generations of selection). The infinitesimal model, which underpins most genetic and genomic evaluations, assumes that each quantitative trait is controlled by an infinitely large number of unlinked and non-epistatic loci, each with an infinitely small effect. Under the infinitesimal model, selection is not expected to noticeably alter the allele frequencies, despite a potential substantial change in the population mean; the exception is in the first few generations of selection when genetic variance is expected to decline, after which it stabilizes. Despite the common use of the heritability statistic in quantitative genetics as a descriptor of adaption or response to selection, it is arguably the coefficient of genetic variation that is more informative to gauge adaptation potential and should, therefore, always be cited in such studies; for example, the heritability of fertility traits in dairy cows is generally low, yet the coefficient of genetic variation for most traits is comparable to many other performance traits, thus supporting the observed rapid genetic gain in fertility performance in dairy populations. Empirical evidence from long-term selection studies, across a range of animal and plant species, fails to support the premise that selection will deplete genetic variability. Even after 100 yr (synonymous with 100 generations) of selection in corn for high protein or oil content, there appears to be no obvious plateauing in the response to selection. Although populations in several selection experiments did reach a selection limit after multiple generations of directional selection, this does not equate to an exhaustion of genetic variance; such a declaration is supported by the observed rapid responses to reverse selection once implemented in long-term selection studies. New technologies such as genome-wide enabled selection and genome editing, as well as having the potential to accelerate genetic gain, could also increase the genetic variation, or at least reduce the erosion of genetic variance over time. In conclusion, there is no evidence, either theoretical or empirical, to indicate that dairy cow breeding programs will be unable to adapt to evolving challenges and opportunities, at least not because of an absence of ample genetic variability.

Artificial Selection to Increase the Phenotypic Variance in gmax Fails

Stabilizing selection is important in evolutionary theories of the maintenance of genetic variance and has been invoked as the key process determining macroevolutionary patterns of trait evolution. However, manipulative evidence for the extent of stabilizing selection, particularly on multivariate traits, is lacking. We used artificial disruptive selection in Drosophila serrata as a tool to determine the relative strength of stabilizing selection experienced by multivariate trait combinations with contrasting levels of genetic and mutational variance. Contrary to expectation, when disruptive selection was applied to the major axis of standing genetic variance, g_max, we observed a significant and repeatable decrease in its phenotypic variance. In contrast, the multivariate trait combination predicted to be under strong stabilizing selection showed a significant and repeatable increase in its phenotypic variance. Correlated responses were observed in all selection treatments, and viability selection operating on extreme phenotypes of traits genetically correlated with those directly selected on limited our ability to increase their phenotypic range. Our manipulation revealed that multivariate trait combinations were subject to stabilizing selection; however, we did not observe a direct relationship between the strength of stabilizing selection and the levels of standing genetic variance in multivariate trait combinations. Contrasting patterns of allele frequencies underlying traits with high versus low levels of standing genetic variance may be implicated in determining the response to artificial selection in multivariate trait combinations.

PERSPECTIVE: THE PACE OF MODERN LIFE: MEASURING RATES OF CONTEMPORARY MICROEVOLUTION

We evaluate methods for measuring and specifying rates of microevolution in the wild, with particular regard to studies of contemporary, often deemed "rapid," evolution. A considerable amount of ambiguity and inconsistency persists within the field, and we provide a number of suggestions that should improve study design, inference, and clarity of presentation. (1) Some studies measure change over time within a population (allochronic) and others measure the difference between two populations that had a common ancestor in the past (synchronic). Allochronic studies can be used to estimate rates of "evolution," whereas synchronic studies more appropriately estimate rates of "divergence." Rates of divergence may range from a small fraction to many times the actual evolutionary rates in the component populations. (2) Some studies measure change using individuals captured from the wild, whereas others measure differences after rearing in a common environment. The first type of study can be used to specify "phenotypic" rates and the later "genetic" rates. (3) The most commonly used evolutionary rate metric, the darwin, has a number of theoretical shortcomings. Studies of microevolution would benefit from specifying rates in standard deviations per generation, the haldane. (4) Evolutionary rates are typically specified without an indication of their precision. Readily available methods for specifying confidence intervals and statistical significance (regression, bootstrapping, randomization) should be implemented. (5) Microevolutionists should strive to accumulate time series, which can reveal temporal shifts in the rate of evolution and can be used to identify evolutionary patterns. (6) Evolutionary rates provide a convenient way to compare the tempo of evolution across studies, traits, taxa, and time scales, but such comparisons are subject to varying degrees of confidence. Comparisons across different time scales are particularly tenuous. (7) A number of multivariate rate measures exist, but considerable theoretical development is required before their utility can be determined. We encourage the continued investigation of evolutionary rates because the information they provide is relevant to a wide range of theoretical and practical issues.

How does epistasis influence the response to selection?

Much of quantitative genetics is based on the 'infinitesimal model', under which selection has a negligible effect on the genetic variance. This is typically justified by assuming a very large number of loci with additive effects. However, it applies even when genes interact, provided that the number of loci is large enough that selection on each of them is weak relative to random drift. In the long term, directional selection will change allele frequencies, but even then, the effects of epistasis on the ultimate change in trait mean due to selection may be modest. Stabilising selection can maintain many traits close to their optima, even when the underlying alleles are weakly selected. However, the number of traits that can be optimised is apparently limited to ~4Ne by the 'drift load', and this is hard to reconcile with the apparent complexity of many organisms. Just as for the mutation load, this limit can be evaded by a particular form of negative epistasis. A more robust limit is set by the variance in reproductive success. This suggests that selection accumulates information most efficiently in the infinitesimal regime, when selection on individual alleles is weak, and comparable with random drift. A review of evidence on selection strength suggests that although most variance in fitness may be because of alleles with large Nes, substantial amounts of adaptation may be because of alleles in the infinitesimal regime, in which epistasis has modest effects.

Obstruction of adaptation in diploids by recessive, strongly deleterious alleles

Recessive deleterious mutations are common, causing many genetic disorders in humans and producing inbreeding depression in the majority of sexually reproducing diploids. The abundance of recessive deleterious mutations in natural populations suggests they are likely to be present on a chromosome when a new adaptive mutation occurs, yet the dynamics of recessive deleterious hitchhikers and their impact on adaptation remains poorly understood. Here we model how a recessive deleterious mutation impacts the fate of a genetically linked dominant beneficial mutation. The frequency trajectory of the adaptive mutation in this case is dramatically altered and results in what we have termed a "staggered sweep." It is named for its three-phased trajectory: (i) Initially, the two linked mutations have a selective advantage while rare and will increase in frequency together, then (ii), at higher frequencies, the recessive hitchhiker is exposed to selection and can cause a balanced state via heterozygote advantage (the staggered phase), and (iii) finally, if recombination unlinks the two mutations, then the beneficial mutation can complete the sweep to fixation. Using both analytics and simulations, we show that strongly deleterious recessive mutations can substantially decrease the probability of fixation for nearby beneficial mutations, thus creating zones in the genome where adaptation is suppressed. These mutations can also significantly prolong the number of generations a beneficial mutation takes to sweep to fixation, and cause the genomic signature of selection to resemble that of soft or partial sweeps. We show that recessive deleterious variation could impact adaptation in humans and Drosophila.

Adaptations to sexual selection and sexual conflict: insights from experimental evolution and artificial selection

Artificial selection and experimental evolution document natural selection under controlled conditions. Collectively, these techniques are continuing to provide fresh and important insights into the genetic basis of evolutionary change, and are now being employed to investigate mating behaviour. Here, we focus on how selection techniques can reveal the genetic basis of post-mating adaptations to sexual selection and sexual conflict. Alteration of the operational sex ratio of adult Drosophila over just a few tens of generations can lead to altered ejaculate allocation patterns and the evolution of resistance in females to the costly effects of elevated mating rates. We provide new data to show how male responses to the presence of rivals can evolve. For several traits, the way in which males responded to rivals was opposite in lines selected for male-biased, as opposed to female-biased, adult sex ratio. This shows that the manipulation of the relative intensity of intra- and inter-sexual selection can lead to replicable and repeatable effects on mating systems, and reveals the potential for significant contemporary evolutionary change. Such studies, with important safeguards, have potential utility for understanding sexual selection and sexual conflict across many taxa. We discuss how artificial selection studies combined with genomics will continue to deepen our knowledge of the evolutionary principles first laid down by Darwin 150 years ago.

Polymorphic genes of major effect: consequences for variation, selection and evolution in Arabidopsis thaliana

The importance of genes of major effect for evolutionary trajectories within and among natural populations has long been the subject of intense debate. For example, if allelic variation at a major-effect locus fundamentally alters the structure of quantitative trait variation, then fixation of a single locus can have rapid and profound effects on the rate or direction of subsequent evolutionary change. Using an Arabidopsis thaliana RIL mapping population, we compare G-matrix structure between lines possessing different alleles at ERECTA, a locus known to affect ecologically relevant variation in plant architecture. We find that the allele present at ERECTA significantly alters G-matrix structure-in particular the genetic correlations between branch number and flowering time traits-and may also modulate the strength of natural selection on these traits. Despite these differences, however, when we extend our analysis to determine how evolution might differ depending on the ERECTA allele, we find that predicted responses to selection are similar. To compare responses to selection between allele classes, we developed a resampling strategy that incorporates uncertainty in estimates of selection that can also be used for statistical comparisons of G matrices.

Long-term selection for a quantitative character in large replicate populations ofDrosophila melanogaster: II. Lethals and visible mutants with large effects

Lethal frequencies on the second and third chromosomes were estimated three times in six replicate lines ofDrosophila melanogasterselected for increased abdominal bristle number, at G 14–16, G 37–44 and G 79. Ten lethals were detected at a frequency of about 5% or higher at G 14–16, of which only one recurred in subsequent tests. Another ten lethals which had not been detected previously were found at G 37–44, and the 5 most frequent ones recurred at G 79. In the last test, 15 presumably new lethals were detected, of which at least 4 appeared well established. In addition, six reversions (fromsctosc+), a new mutant at the scute locus andscawere discovered. The effects on the selected character of some lethals and visible mutants were large and variable, but not always sufficient to explain the observed frequencies. The major lethals detected at G 37–44 and G 79 for the first time were most probably ‘mutations’ (in the broad sense) which occurred during selection. The likely origins of such ‘mutations’ were discussed, with a suggestion that the known mutation rate for recessive lethals would not be incompatible with the observed frequency of occurrence of the ‘mutations’.

Artificial selection with differing population structures

The effect of subdivision of a population on response to artificial directional selection for abdominal bristle number inDrosophila melanogasterwas compared using large, replicated lines. Three different population structures were compared: (i) selection in an Undivided, large population with 50 pairs of parents (treatment U); (ii) selection in each of 10 sublines which were reconstituted every 6th generation by Crossing after Culling the 5 lowest sublines (treatment CC); and (iii) selection in each of 10 sublines which were reconstituted every 6th generation by Crossing after Retaining all 10 sublines (treatment CR). At the end of three cycles of selection and crossing, neither CR nor CC was superior to U; sublining did not increase response to selection. These results agree with the predictions arising from an entirely additive model and provide no evidence for the presence of epistasis.

A comparison of 50-pair lines (U) with several 5-pair lines was made over 31 generations. For the 50-pair lines, there was close agreement between response predicted from the base population (usingih2σp) and observed response throughout all 31 generations of selection. Although the best of the 5-pair lines exceeded the 50-pair lines in the early generations, average response to directional selection in the 5-pair lines soon fell behind that predicted fromih2σp, and soon reached a plateau.

Theoretical models of selection and mutation on quantitative traits

Empirical studies of quantitative genetic variation have revealed robust patterns that are observed both across traits and across species. However, these patterns have no compelling explanation, and some of the observations even appear to be mutually incompatible. We review and extend a major class of theoretical models, 'mutation-selection models', that have been proposed to explain quantitative genetic variation. We also briefly review an alternative class of 'balancing selection models'. We consider to what extent the models are compatible with the general observations, and argue that a key issue is understanding and modelling pleiotropy. We discuss some of the thorny issues that arise when formulating models that describe many traits simultaneously.

Predictions of patterns of response to artificial selection in lines derived from natural populations

The pattern of response to artificial selection on quantitative traits in laboratory populations can tell us something of the genetic architecture in the natural population from which they were derived. We modeled artificial selection in samples drawn from natural populations in which variation had been maintained by recurrent mutation, with genes having an effect on the trait, which was subject to real stabilizing selection, and a pleitropic effect on fitness (the joint-effect model). Natural selection leads to an inverse correlation between effects and frequencies of genes, such that the frequency distribution of genes increasing the trait has an extreme U-shape. In contrast to the classical infinitesimal model, an early accelerated response and a larger variance of response among replicates were predicted. However, these are reduced if the base population has been maintained in the laboratory for some generations by random sampling prior to artificial selection. When multiple loci and linkage are also taken into account, the gametic disequilibria generated by the Bulmer and Hill-Robertson effects are such that little or no increase in variance and acceleration of response in early generations of artificial selection are predicted; further, the patterns of predicted responses for the joint-effect model now become close to those of the infinitesimal model. Comparison with data from laboratory selection experiments shows that, overall, the analysis did not provide clear support for the joint-effect model or a clear case for rejection.

Patterns of genetic variation in Mendelian and complex traits

This review discusses the prospects for understanding the genetic basis of complex traits in humans. We take the view that work done on Drosophila melanogaster can serve as a model for understanding complex traits in humans, and the literature on this model system, as well as on humans, is reviewed. The prospects for success in understanding the genetic basis of complex traits depend, in part, on the nature of the forces acting on genetic variation. We suggest that different experimental approaches should be undertaken for traits caused by common genetic variants versus those arising from rare genetic variants.

Understanding quantitative genetic variation

Until recently, it was impracticable to identify the genes that are responsible for variation in continuous traits, or to directly observe the effects of their different alleles. Now, the abundance of genetic markers has made it possible to identify quantitative trait loci (QTL)--the regions of a chromosome or, ideally, individual sequence variants that are responsible for trait variation. What kind of QTL do we expect to find and what can our observations of QTL tell us about how organisms evolve? The key to understanding the evolutionary significance of QTL is to understand the nature of inherited variation, not in the immediate mechanistic sense of how genes influence phenotype, but, rather, to know what evolutionary forces maintain genetic variability.

Comparison of genetic variation in growth performance on normal and novel host plants in a local population of a herbivorous ladybird beetle, Epilachna vigintioctomaculata

Local populations of herbivore species that display variation in host plant use are subject to natural selection for improved ability to use their own host species. Since natural selection changes the frequency of alleles that control host use, genetic variation in growth performance on host plants may change with each generation of selection. Therefore, within-population variations in the ability to use different hosts may reflect past selective forces. The present study reports the genetic variation in growth performance in a population of the herbivorous ladybird beetle, Epilachna vigintioctomaculata Motschulsky (Coccinellidae, Epilachninae) on both the normal host and a novel host. The present study found higher heritabilities for growth performance on the novel host than the normal host. The difference in heritabilities was caused by smaller among-family variance components on the normal host compared to those on the novel host. The results are compatible with the view that natural selection, which improved the ability of the population to use a particular host plant, has reduced genetic variation within that population in the ability to use that host plant. The pattern of reduced genetic variation in a population reared on the normal host observed in the present study is compared with that in another population, which showed local adaptation in the use of its own host plant. The effect of natural selection on differing host plant use on within-population genetic variation is discussed.

Laboratory selection experiments using Drosophila: what do they really tell us?

Laboratory selection experiments using Drosophila, and other organisms, are widely used in experimental biology. In particular, such experiments on D. melanogaster life history and stress-related traits have been instrumental in developing the emerging field of experimental evolution. However, similar selection experiments often produce inconsistent correlated responses to selection. Unfortunately, selection experiments are vulnerable to artifacts that are difficult to control. In spite of these problems, selection experiments are a valuable research tool and can contribute to our understanding of evolution in natural populations.

Candidate quantitative trait loci and naturally occurring phenotypic variation for bristle number in Drosophila melanogaster: the Delta-Hairless gene region

Delta (Dl) and Hairless (H) are two chromosome 3 candidate neurogenic loci that might contribute to naturally occurring quantitative variation for sensory bristle number. To evaluate this hypothesis, we assessed quantitative genetic variation in abdominal and sternopleural bristle numbers among homozygous isogenic third chromosomes sampled from nature and substituted into the Samarkand (Sam) inbred chromosome 1 and 2 background; among homozygous lines in which the wild-derived Dl-H gene region was introgressed into the Sam chromosome 3 background; and among Dl-H region introgression lines as heterozygotes against the Sam wild-type strain and derivatives of Sam into which mutant Dl and H alleles had been introgressed. Variation among the Dl-H region introgression lines accounted for 36% (8.3%) of the total chromosome 3 among line variance in abdominal (sternopleural) bristle number and for 53% of the chromosome 3 sex x line variance in abdominal bristle number. Naturally occurring alleles in the Dl-H region failed to complement a Dl mutant allele for female abdominal bristle number and sternopleural bristle number in both sexes, and an H mutant allele for both bristle traits in males and females. These results are consistent with the hypothesis that naturally occurring alleles at Dl and H contribute to quantitative genetic variation in sensory bristle number.

The tumorous-head-1 locus affects bristle number of the Drosophila melanogaster cuticle

The tuh-1 maternal effect locus contains two naturally occurring isoalleles, tuh-1h and tuh-1g. Until recently there has been no possibility to distinguish between the tuh-lh and the tuh-1g maternal effects other than evaluating their effect on the Bithorax-Complex (BXC) Abdominal B (Abd-B) mutant tuh-3. However, in this report we identify a bristle phenotype associated with the tuh-1 locus that has very interesting evolutionary implications. Females homozygous for tuh-1h always produce adult offspring with more bristles than females homozygous or heterozygous for tuh-1g. The effect is global. Increased bristle number occurs in the head, the thorax, and the anterior and posterior abdomen. Females totally deficient for the tuh-1 gene produce offspring with high bristle number. Thus, the bristle phenotype results from the absence of the maternally contributed tuh-1g factor. Genetic evidence shows that the bristle phenotype is caused by the tuh-1 locus and that tuh-1h is completely recessive to tuh-1g. The tuh-1 locus is located at the euchromatin-beta-heterochromatin junction near the centromere of the X chromosome and deficiency analysis places the locus between the lethal genes extra organs (eo) and lethal B20 (lB20). The variance in bristle number attributable to the tuh-1 locus in nature is approximately 10.1%, an indication that the bristle phenotype is most likely a neutral, pleiotrophic side effect of tuh-1.

Adaptation and extinction in changing environments

The extinction risk of a population is determined by its demographic properties, the environmental conditions to which it is exposed, and its genetic potential to cope with and adapt to its environment. All these factors may have stochastic as well as directional components. The present chapter reviews several types of models concerned with the vulnerability of small populations to demographic stochasticity and to random and directional changes of the environment. In particular, the influence of mutation and genetic variability on the persistence time of a population is explored, critical rates for environmental change are estimated beyond which extinction on time scales of tens to a few thousand generations is virtually certain, and the extinction risks caused by the above mentioned factors are compared.

A computer simulation evaluation of the role of mutations in finite populations on the response to directional selection: The generations required to attain maximum genetic variance

The role of mutations in finite populations on response to artificial selection was investigated by a computer simulation model designed to mimic the biological model of pupal weight of Tribolium. Given the model, the results showed that with selection about 25-55 generations were needed for genetic variances to reach a maximum value depending on population size, selection intensity, and gene number. When effective population size was larger than 40 or the intensity of selection was high (less than 50% selected), selection had a dramatic effect in reducing the time to approach the maximum point of genetic variance. Furthermore, the genetic variance after that point often declined as a function of selection instead of remaining at a steady state in the subsequent generations.

The causes of extinction

A species may go extinct either because it is unable to evolve rapidly enough to meet changing circumstances, or because its niche disappears and no capacity for rapid evolution could have saved it. Although recent extinctions can usually be interpreted as resulting from niche disappearance, the taxonomic distribution of parthenogens suggests that inability to evolve may also be important. A second distinction is between physical and biotic causes of extinction. Fossil evidence for constant taxonomic diversity, combined with species turnover, implies that biotic factors have been important. A similar conclusion emerges from studies of recent introductions of predators, competitors and parasites into new areas. The term 'species selection' should be confined to cases in which the outcome of selection is determined by properties of the population as a whole, rather than of individuals. The process has been of only trivial importance in producing complex adaptations, but of major importance in determining the distribution of different types of organisms. An adequate interpretation of the fossil record requires a theory of the coevolution of many interacting species. Such a theory is at present lacking, but various approaches to it are discussed.

Quantitative genetic variability maintained by mutation-stabilizing selection balance: sampling variation and response to subsequent directional selection

A model of genetic variation of a quantitative character subject to the simultaneous effects of mutation, selection and drift is investigated. Predictions are obtained for the variance of the genetic variance among independent lines at equilibrium with stabilizing selection. These indicate that the coefficient of variation of the genetic variance among lines is relatively insensitive to the strength of stabilizing selection on the character. The effects on the genetic variance of a change of mode of selection from stabilizing to directional selection are investigated. This is intended to model directional selection of a character in a sample of individuals from a natural or long-established cage population. The pattern of change of variance from directional selection is strongly influenced by the strengths of selection at individual loci in relation to effective population size before and after the change of regime. Patterns of change of variance and selection responses from Monte Carlo simulation are compared to selection responses observed in experiments. These indicate that changes in variance with directional selection are not very different from those due to drift alone in the experiments, and do not necessarily give information on the presence of stabilizing selection or its strength.

Time of detection of recessive genes: effects of system of mating and number of examined individuals

The elapsed time from the appearance of a single copy of a recessive gene in a population with separate sexes until the first detection of a recessive homozygote has been analyzed using simulation techniques. Several systems of mating frequently used in laboratory maintenance and artificial selection processes have been the subject of this study. The expected time of detection (T) in each system of mating is described by a function of number of parents (N) and the factor K that when multiplied by N gives the number of examined individuals per generation (e.g. T=1.6+1.8 (N/K) (1/3) for the mass-mating system). Although the expected time of detection changes greatly depending on the system of mating and on the K value (the causes are discussed), both the time scale (N (1/3)) and the coefficient of variation of the distribution of the times of detection (2/3) seem to be unaffected. The additional reduction of the effective population number caused by artificial selection on a heritable trait modifies the expected detection time in a minor way unless both high heritability and high selection intensity are involved.

Mutational variance for pupa weight in Tribolium castaneum

Directional selection for heavier pupa weight in Tribolium castaneum was practiced for 18 generations in two replicates of an inbred line, each separately maintained in small population cages for more than 90 generations. Mutational variance was estimated in two ways, based on Hill's (1982a) prediction equation for response to directional selection where an equilibrium state between effective population size and variation created by new mutation is assumed. Estimates of mutational variance based on response to selection in a selected population and from a sire-offspring regression analysis in an unselected control population were in strong agreement within each replicate population. Significant differences between the two replicates were observed. Estimates of the ratio of mutational variance to environmental variance ranged from 0.0002 to 0.0012, depending upon the assumptions made about effective population sizes maintained in the two replicate lines. Estimates of realized heritability from the 18 generations of selection were 0.23±0.02 and 0.12±0.02 in the two replicates. The results support the hypothesis that mutation may have played a significant role in supplying useful genetic variation for long-continuing response to selection for this trait in experiments reported earlier.

Genetic divergence under uniform selection. III. Selection for knockdown resistance to ethanol in Drosophila pseudoobscura populations and their replicate lines

Replicate D. pseudoobscura lines from populations collected at different geographic locations were selected for increased knockdown resistance to ethanol. Population background affected the initial rate of response but not the extent that lines responded. Lines were tested for physiological traits contributing to increased knockdown resistance. Populations showed different correlated responses for two traits (tolerance of ethanol, and of acetone), suggesting that they had responded to selection by different mechanisms. Replicate lines had diverged for most traits. The results indicate that drift and/or differences in genetic background can lead to divergence under uniform selection, even when fairly large population sizes are maintained.

Quantitative genetics and fitness: lessons from Drosophila

This paper examines patterns of heritability and genetic covariance between traits in the genus Drosophila. Traits are divided into the categories, morphology, behaviour, physiology and life history. Early theoretical analyses suggested that life history traits should have heritabilities that are lower than those in other categories. Variable pleiotrophy, environmental variation, mutation and niche variation may, however, maintain high heritabilities. In Drosophila the heritabilities of life history traits are lower than morphological or physiological traits but may exceed 20 per cent. The pattern of variation in the heritability of behavioural traits is similar to that of life history traits. Genetic covariance between morphological traits and between morphological and life history traits are all positive but those between life history traits have variable sign. Negative covariance between traits supports the variable pleiotropy hypothesis but other factors such as environmental heterogeneity, or mutation cannot be excluded.

Models of long-term artificial selection in finite population with recurrent mutation

The effects of mutation on mean and variance of response to selection for quantitative traits are investigated. The mutants are assumed to be unlinked, to be additive, and to have their effects symmetrically distributed about zero, with absolute values of effects having a gamma distribution. It is shown that the ratio of expected cumulative response to generation t from mutants, , and expected response over one generation from one generation of mutants, , is a function of t/N, where t is generations and N is effective population size. Similarly, , is a function of t/N, where is the increment in genetic variance from one generation of mutants. The mean and standard deviation of response from mutations relative to that from initial variation in the population, in the first generation, are functions of . Evaluation of these formulae for a range of parameters quantifies the important role that population size can play in response to long-term selection.

A selection response plateau for radiation resistance in the cotton boll weevil

Twenty generations of family selection in the cotton boll weevil for 14-day postirradiation survival to 10,000 rads of gamma irradiation has increased survival to nearly 90% as compared with about 35% in the unselected control population. Mean survival time has increased to 21.2 days in the selected population, as compared with 12.8 days in the unselected control. Nearly all of the response to selection occurred in the first 12 generations of selection, with no significant improvement beyond that point. A relaxed selection line was established in generation 12 and has been maintained as a population cage with discrete generations since that time. A comparison in generation 17 between the relaxed selection population and the selected population where both populations were managed in the same way indicated that none of the increase in resistance had been lost due to relaxation of selection (89.2% survival in the relaxed population as compared with 86.0% in the selected population). The rapid increase in response to selection followed by a quick plateau and no decline in the mean following relaxation of selection support the hypothesis that the increased resistance to irradiation resulted from changes in allelic frequencies for a relatively small number of genes. Alleles for increased resistance were either fixed by the selection process or, if still segregating, were not negatively correlated with fitness. Estimates of heritability for other fitness traits indicate selection should be effective for several other traits of importance in the efficiency of a mass rearing program.

Predictions of response to artificial selection from new mutations

The pattern of response expected from fixation of mutant genes for quantitative traits in finite populations is investigated for a range of distributions of mutant gene effects. The eventual rate depends on the total variance of mutant effects per generation, but the initial rate and the variance of response is higher if the distribution of mutant effects has a large standard deviation or is leptokurtic. The difference between initial and eventual rates of response is greater with large population sizes.

For a range of assumptions, new mutants are unlikely to have much influence on response for 20 or so generations, but then may contribute substantially, such that no plateaux are obtained. However, information on the variance contributed by mutants is almost entirely on bristle number inDrosophila.

It is argued that the role of new mutants should be considered in designing breeding programmes, in particular in utilizing large populations.

Effect of short term directional selection on genetic variability: experiments with Drosophila melanogaster

Experimental checks on theoretical predictions of the build up of negative linkage disequilibrium with directional selection were made using abdominal bristle number in Drosophila melanogaster. Selection was practised for three generations before relaxation. Realized heritabilities and thus genotype variances were estimated by divergent selection. In one replicate, little change of variance occurred but in the other it increased substantially on relaxation. This result is compatible with a model of one or more genes of large effect at extreme frequencies in the base population. This is illustrated with Monte Carlo simulations. Interpretations of results was aided by considering the build up of negative disequilibrium.