Evidence for adaptive phenotypic differentiation in Baltic Sea sticklebacks

The evidence for adaptive phenotypic differentiation in mobile marine species remains scarce, partly due to the difficulty of obtaining quantitative genetic data to demonstrate the genetic basis of the observed phenotypic differentiation. Using a combination of phenotypic and molecular genetic approaches, we elucidated the relative roles of natural selection and genetic drift in explaining lateral plate number differentiation in threespine sticklebacks (Gasterosteus aculeatus) across the entire Baltic Sea basin (approximately 392 000 km²). We found that phenotypic differentiation (P_ST = 0.213) in plate number exceeded that in neutral markers (F_ST = 0.008), suggesting an adaptive basis for the observed differentiation. Because a close correspondence was found between plate phenotype and genotype at a quantitative trait loci (QTL; STN381) tightly linked to the gene (Ectodysplasin) underlying plate variation, the evidence for adaptive differentiation was confirmed by comparison of F_ST at the QTL (F_STQ = 0.089) with F_ST at neutral marker loci. Hence, the results provide a comprehensive demonstration of adaptive phenotypic differentiation in a high‐gene‐flow marine environment with direct, rather than inferred, verification for the genetic basis of this differentiation. In general, the results illustrate the utility of P_ST–F_ST–F_STQ comparisons in uncovering footprints of natural selection and evolution and add to the growing evidence for adaptive genetic differentiation in high‐gene‐flow marine environments, including that of the relatively young Baltic Sea.     

Are QST–FST comparisons for natural populations meaningful?

Comparisons between putatively neutral genetic differentiation amongst populations, F_ST, and quantitative genetic variation, Q_ST, are increasingly being used to test for natural selection. However, we find that approximately half of the comparisons that use only data from wild populations confound phenotypic and genetic variation. We urge the use of a clear distinction between narrow‐sense Q_ST, which can be meaningfully compared with F_ST, and phenotypic divergence measured between populations, P_ST, which is inadequate for comparisons in the wild. We also point out that an unbiased estimate of Q_ST can be found using the so‐called ‘animal model’ of quantitative genetics.     

Local selection modifies phenotypic divergence among Rana temporaria populations in the presence of gene flow

In ectotherms, variation in life history traits among populations is common and suggests local adaptation. However, geographic variation itself is not a proof for local adaptation, as genetic drift and gene flow may also shape patterns of quantitative variation. We studied local and regional variation in means and phenotypic plasticity of larval life history traits in the common frog Rana temporaria using six populations from central Sweden, breeding in either open‐canopy or partially closed‐canopy ponds. To separate local adaptation from genetic drift, we compared differentiation in quantitative genetic traits (Q_ST) obtained from a common garden experiment with differentiation in presumably neutral microsatellite markers (F_ST). We found that R. temporaria populations differ in means and plasticities of life history traits in different temperatures at local, and in F_ST at regional scale. Comparisons of differentiation in quantitative traits and in molecular markers suggested that natural selection was responsible for the divergence in growth and development rates as well as in temperature‐induced plasticity, indicating local adaptation. However, at low temperature, the role of genetic drift could not be separated from selection. Phenotypes were correlated with forest canopy closure, but not with geographical or genetic distance. These results indicate that local adaptation can evolve in the presence of ongoing gene flow among the populations, and that natural selection is strong in this system.     

Comparison of genetic differentiation at marker loci and quantitative traits

The comparison of the degree of differentiation in neutral marker loci and genes coding quantitative traits with standardized and equivalent measures of genetic differentiation (F_ST and Q_ST, respectively) can provide insights into two important but seldom explored questions in evolutionary genetics: (i) what is the relative importance of random genetic drift and directional natural selection as causes of population differentiation in quantitative traits, and (ii) does the degree of divergence in neutral marker loci predict the degree of divergence in genes coding quantitative traits? Examination of data from 18 independent studies of plants and animals using both standard statistical and meta‐analytical methods revealed a number of interesting points. First, the degree of differentiation in quantitative traits (Q_ST) typically exceeds that observed in neutral marker genes (F_ST), suggesting a prominent role for natural selection in accounting for patterns of quantitative trait differentiation among contemporary populations. Second, the F_ST – Q_ST difference is more pronounced for allozyme markers and morphological traits, than for other kinds of molecular markers and life‐history traits. Third, very few studies reveal situations were Q_ST < F_ST, suggesting that selection pressures, and hence optimal phenotypes, in different populations of the same species are unlikely to be often similar. Fourth, there is a strong correlation between Q_ST and F_ST indices across the different studies for allozyme (r=0.81), microsatellite (r=0.87) and combined (r=0.75) marker data, suggesting that the degree of genetic differentiation in neutral marker loci is closely predictive of the degree of differentiation in loci coding quantitative traits. However, these interpretations are subject to a number of assumptions about the data and methods used to derive the estimates of population differentiation in the two sets of traits.     

Genetic variation in the susceptibility of Eucalyptus globulus to drought damage

Genetic variation in drought damage in Eucalyptus globulus was studied in a sublined trial series across four neighbouring sites in Western Australia linked by ten common families. The trials included approximately 400 open-pollinated families, encompassing 51 native stand collection localities and 19 subraces from throughout the geographic range of the species. Data were analysed using mixed models, with spatial analysis used to better identify genetic effects. Significant subrace differences in drought damage were detected, with both broad-scale, regional and localised clines evident. The quantitative genetic differentiation between subraces as measured by Q _ST (0.39?±?0.091) was significantly greater than the F _ST for neutral marker expectations and consistent with diversifying selection shaping the patterns of subrace divergence in drought susceptibility. This conclusion is supported by the significant association of subrace drought susceptibility with bioclimatic parameters, particularly those associated with temperature seasonality. Less drought damage was observed in subraces originating from areas with more temperature seasonality, but also less radiation and rainfall seasonality, less winter rainfall, higher radiation and higher temperatures in the warmest month. Significant additive genetic variation in drought damage was detected within subraces, with narrow-sense heritabilities ranging from 0.14 to 0.20. We argue that spatial genetic variation in drought susceptibility of E. globulus has been shaped by natural selection acting at multiple scales and discuss opportunities for exploiting this genetic variation in breeding and deployment programs.     

Molecular and quantitative trait variation within and among small fragmented populations of the endangered plant species Psilopeganum sinense

Background and Aims

Natural selection and genetic drift are important evolutionary forces in determining genetic and phenotypic differentiation in plant populations. The extent to which these two distinct evolutionary forces affect locally adaptive quantitative traits has been well studied in common plant and animal species. However, we know less about how quantitative traits respond to selection pressures and drift in endangered species that have small population sizes and fragmented distributions. To address this question, this study assessed the relative strengths of selection and genetic drift in shaping population differentiation of phenotypic traits in Psilopeganum sinense, a naturally rare and recently endangered plant species.

Methods

Population differentiation at five quantitative traits (Q_ST) obtained from a common garden experiment was compared with differentiation at putatively neutral microsatellite markers (F_ST) in seven populations of P. sinense. Q_ST estimates were derived using a Bayesian hierarchical variance component method.

Key Results

Trait-specific Q_ST values were equal to or lower than F_ST. Neutral genetic diversity was not correlated with quantitative genetic variation within the populations of P. sinense.

Conclusions

Despite the prevalent empirical evidence for Q_ST > F_ST, the results instead suggest a definitive role of stabilizing selection and drift leading to phenotypic differentiation among small populations. Three traits exhibited a significantly lower Q_ST relative to F_ST, suggesting that populations of P. sinense might have experienced stabilizing selection for the same optimal phenotypes despite large geographical distances between populations and habitat fragmentation. For the other two traits, Q_ST estimates were of the same magnitude as F_ST, indicating that divergence in these traits could have been achieved by genetic drift alone. The lack of correlation between molecular marker and quantitative genetic variation suggests that sophisticated considerations are required for the inference of conservation measures of P. sinense from neutral genetic markers.     

Phenotypic divergence of the common toad (Bufo bufo) along an altitudinal gradient: evidence for local adaptation

Variation in the environment can induce different patterns of genetic and phenotypic differentiation among populations. Both neutral processes and selection can influence phenotypic differentiation. Altitudinal phenotypic variation is of particular interest in disentangling the interplay between neutral processes and selection in the dynamics of local adaptation processes but remains little explored. We conducted a common garden experiment to study the phenotypic divergence in larval life-history traits among nine populations of the common toad (Bufo bufo) along an altitudinal gradient in France. We further used correlation among population pairwise estimates of quantitative trait (Q_ST) and neutral genetic divergence (F_ST from neutral microsatellite markers), as well as altitudinal difference, to estimate the relative role of divergent selection and neutral genetic processes in phenotypic divergence. We provided evidence for a neutral genetic differentiation resulting from both isolation by distance and difference in altitude. We found evidence for phenotypic divergence along the altitudinal gradient (faster development, lower growth rate and smaller metamorphic size). The correlation between pairwise Q_STs–F_STs and altitude differences suggested that this phenotypic differentiation was most likely driven by altitude-mediated selection rather than by neutral genetic processes. Moreover, we found different divergence patterns for larval traits, suggesting that different selective agents may act on these traits and/or selection on one trait may constrain the evolution on another through genetic correlation. Our study highlighted the need to design more integrative studies on the common toad to unravel the underlying processes of phenotypic divergence and its selective agents in the context of environmental clines.     

Genetic drift and uniform selection shape evolution of most traits in <Emphasis Type="Italic">Eugenia dysenterica</Emphasis> DC. (Myrtaceae)

Knowing how microevolutionary processes, such as genetic drift and natural selection, shape variation in adaptive traits is strategic for conservation measures. One way to estimate local adaptation is to compare divergences in quantitative traits (Q_ST) and neutral loci (F_ST). Therefore, we have assessed the pattern of phenotypic and molecular genetic divergence among natural subpopulations of the fruit tree Eugenia dysenterica DC. A provenance and progeny test was performed to assess the quantitative traits of the subpopulations collected in a wide distribution area of the species in the Brazilian Cerrado. The sampled environments are in a biodiversity hotspot with heterogeneous soil and climate conditions. By associating quantitative trait variation in initial seedling development with neutral microsatellite marker variation, we tested the local adaptation of the traits by the Q_ST–F_ST contrast. Genetic drift was prevalent in the phenotypic differentiation among the subpopulations, although the traits seedling emergence time and root green mass, which are relevant for adaptation to the Cerrado climate, showed signs of uniform selection. Our results suggest that E. dysenterica has a spatial genetic structure divided into two large groups, separated by a line that divides the Cerrado biome in a southwestern to northeastern direction. This structure must be taken into account for managing E. dysenterica genetic resources both for conservation and breeding purposes.     

Morphometric differentiation across House Sparrow Passer domesticus populations in Finland in comparison with the neutral expectation for divergence

To understand the biology of organisms it is important to take into account the evolutionary forces that have acted on their constituent populations. Neutral genetic variation is often assumed to reflect variation in quantitative traits under selection, though with even low neutral divergence there can be substantial differentiation in quantitative genetic variation associated with locally adapted phenotypes. To study the relative roles of natural selection and genetic drift in shaping phenotypic variation, the levels of quantitative divergence based on phenotypes (P_ST) and neutral genetic divergence (F_ST) can be compared. Such a comparison was made between 10 populations of Finnish House Sparrows (n = 238 individuals) collected in 2009 across the whole country. Phenotypic variation in tarsus‐length, wing‐length, bill‐depth, bill‐length and body mass were considered and 13 polymorphic microsatellite loci were analysed to quantify neutral genetic variation. Calculations of P_ST were based on Markov‐Chain Monte Carlo Bayesian estimates of phenotypic variances across and within populations. The robustness of the conclusions of the P_ST–F_ST comparison was evaluated by varying the proportion of variation due to additive genetic effects within and across populations. Our results suggest that body mass is under directional selection, whereas the divergence in other traits does not differ from neutral expectations. These findings suggest candidate traits for considering gene‐based studies of local adaptation. The recognition of locally adapted populations may be of value in the conservation of this declining species.     

Natural selection contributes to geographic patterns of thermal plasticity in Plantago lanceolata

A long‐standing debate in evolutionary biology concerns the relative importance of different evolutionary forces in explaining phenotypic diversification at large geographic scales. For example, natural selection is typically assumed to underlie divergence along environmental gradients. However, neutral evolutionary processes can produce similar patterns. We collected molecular genetic data from 14 European populations of Plantago lanceolata to test the contributions of natural selection versus neutral evolution to population divergence in temperature‐sensitive phenotypic plasticity of floral reflectance. In P. lanceolata, reflectance plasticity is positively correlated with latitude/altitude. We used population pairwise comparisons between neutral genetic differentiation (F_ST and Jost's D) and phenotypic differentiation (P_ST) to assess the contributions of geographic distance and environmental parameters of the reproductive season in driving population divergence. Data are consistent with selection having shaped large‐scale geographic patterns in thermal plasticity. The aggregate pattern of P_ST versus F_ST was consistent with divergent selection. F_ST explained thermal plasticity differences only when geographic distance was not included in the model. Differences in the extent of cool reproductive season temperatures, and not overall temperature variation, explained plasticity differences independent of distance. Results are consistent with the hypothesis that thermal plasticity is adaptive where growing seasons are shorter and cooler, that is, at high latitude/altitude.     

Multiple evolutionary processes drive the patterns of genetic differentiation in a forest tree species complex

Forest trees frequently form species complexes, complicating taxonomic classification and gene pool management. This is certainly the case in Eucalyptus, and well exemplified by the Eucalyptus globulus complex. This ecologically and economically significant complex comprises four taxa (sspp. bicostata, globulus, maidenii, pseudoglobulus) that are geographically and morphologically distinct, but linked by extensive “intergrade” populations. To resolve their genetic affinities, nine microsatellites were used to genotype 1200 trees from throughout the natural range of the complex in Australia, representing 33 morphological core and intergrade populations. There was significant spatial genetic structure (F_ST = 0.10), but variation was continuous. High genetic diversity in southern ssp. maidenii indicates that this region is the center of origin. Genetic diversity decreases and population differentiation increases with distance from this area, suggesting that drift is a major evolutionary process. Many of the intergrade populations, along with other populations morphologically classified as ssp. pseudoglobulus or ssp. globulus, belong to a “cryptic genetic entity” that is genetically and geographically intermediate between core ssp. bicostata, ssp. maidenii, and ssp. globulus. Geography, rather than morphology, therefore, is the best predictor of overall genetic affinities within the complex and should be used to classify germplasm into management units for conservation and breeding purposes.     

Contrasting Levels of Variation in Neutral and Quantitative Genetic Loci on Island Populations of Moor Frogs (<Emphasis Type="Italic">Rana arvalis</Emphasis>)

Reduced levels of genetic variability and a prominent differentiation in both neutral marker genes and phenotypic traits are typical for many island populations as compared to their mainland conspecifics. However, whether genetic diversity in neutral marker genes reflects genetic variability in quantitative traits, and thus, their evolutionary potential, remains typically unclear. Moreover, the phenotypic differentiation on islands could be attributable to phenotypic plasticity, selection or drift; something which seldom has been tested. Using eight polymorphic microsatellite loci and quantitative genetic breeding experiments we conducted a detailed comparison on genetic variability and differentiation between Nordic islands (viz. Gotland, Öland and Læsø) and neighbouring mainland populations of moor frogs (Rana arvalis). As expected, the neutral variation was generally lower in island than in mainland populations. But as opposed to this, higher levels of additive genetic variation (V _A) in body size and tibia length were found on the island of Gotland as compared to the mainland population. When comparing the differentiation seen in neutral marker genes (F _ST) with the differentiation in genes coding quantitative traits (Q _ST) two different evolutionary scenarios were found: while selection might explain a smaller size of moor frogs on Gotland, the differentiation seen in tibia length could be explained by genetic drift. These results highlight the limited utility of microsatellite loci alone in inferring the causes behind an observed phenotypic differentiation, or in predicting the amount of genetic variation in ecologically important quantitative traits.     

The importance of phenotypic differentiation and plasticity in the range expansion of the annual Atriplex tatarica (Amaranthaceae)

The ability of a species to adapt to sub-optimal conditions at the margin of its distribution range and to cope with environmental stress is considered to be important for its successful geographic expansion. To ascertain the roles of phenotypic differentiation and plasticity in the expansion of the annual Atriplex tatarica, we compared plants from populations found in Marginal and Central areas of the species’ range. We grew these plants under marginal climatic conditions in pots with different types of substrate. We assessed the population genetic structure at five putatively neutral allozyme loci to evaluate whether there was any evidence of reduced genetic diversity in Marginal populations compared to Central ones. We used the Q_ST vs. F_ST approach (while F_ST gives a standardised measure of the genetic differentiation among populations for a genetic locus, Q_ST measures the amount of genetic variance among populations relative to the total genetic variance) to ascertain the roles of adaptive vs. non-adaptive processes on phenotypic differentiation. Plants native to the Marginal area of the species’ range flowered earlier and had a lower shoot mass and a higher reproductive allocation than plants native to the Central part of the species’ range. The Marginal populations of Atriplex tatarica showed lower genetic diversity at allozyme loci and higher phenotypic differentiation than the Central populations. We recorded similar plastic responses to substrates in plants native to both regions. Our results indicate that Marginal populations of expanding A. tatarica maintain the ability to adapt locally and to elicit a plastic response to environmental stress, despite loss of genetic diversity.     

QST–FST comparisons with unbalanced half‐sib designs

Q_ST, a measure of quantitative genetic differentiation among populations, is an index that can suggest local adaptation if Q_ST for a trait is sufficiently larger than the mean F_ST of neutral genetic markers. A previous method by Whitlock and Guillaume derived a simulation resampling approach to statistically test for a difference between Q_ST and F_ST, but that method is limited to balanced data sets with offspring related as half‐sibs through shared fathers. We extend this approach (i) to allow for a model more suitable for some plant populations or breeding designs in which offspring are related through mothers (assuming independent fathers for each offspring; half‐sibs by dam); and (ii) by explicitly allowing for unbalanced data sets. The resulting approach is made available through the R package QstFstComp.     

The effects of selection,drift and genetic variation on life‐history trait divergence among insular populations of natterjack toad,Bufo calamita

Although loss of genetic variation is frequently assumed to be associated with loss of adaptive potential, only few studies have examined adaptation in populations with little genetic variation. On the Swedish west coast, the northern fringe populations of the natterjack toad Bufo calamita inhabit an atypical habitat consisting of offshore rock islands. There are strong among‐population differences in the amount of neutral genetic variation, making this system suitable for studies on mechanisms of trait divergence along a gradient of within‐population genetic variation. In this study, we examined the mechanisms of population divergence using Q_ST–F_ST comparisons and correlations between quantitative and neutral genetic variation. Our results suggest drift or weak stabilizing selection across the six populations included in this study, as indicated by low Q_ST–F_ST values, lack of significant population × temperature interactions and lack of significant differences among the islands in breeding pond size. The six populations included in this study differed in both neutral and quantitative genetic variation. Also, the correlations between neutral and quantitative genetic variation tended to be positive, however, the relatively small number of populations prevents any strong conclusions based on these correlations. Contrary to the majority of Q_ST–F_ST comparisons, our results suggest drift or weak stabilizing selection across the examined populations. Furthermore, the low heritability of fitness‐related traits may limit evolutionary responses in some of the populations.     

Adaptive differentiation in floral traits in the presence of high gene flow in scarlet gilia (Ipomopsis aggregata)

Plant–pollinator interactions are thought to be major drivers of floral trait diversity. However, the relative importance of divergent pollinator‐mediated selection vs. neutral processes in floral character evolution has rarely been explored. We tested for adaptive floral trait evolution by comparing differentiation at neutral genetic loci to differentiation at quantitative floral traits in a putative Ipomopsis aggregata hybrid zone. Typical I. aggregata subsp. candida displays slender white tubular flowers that are typical of flowers pollinated by hawkmoths, and subsp. collina displays robust red tubular flowers typical of flowers pollinated by hummingbirds; yet, hybrid flower morphs are abundant across the East Slope of the Colorado Rockies. We estimated genetic differentiation (F_ST) for nuclear and chloroplast microsatellite loci and used a half‐sib design to calculate quantitative trait divergence (Q_ST) from collection sites across the morphological hybrid zone. We found little evidence for population structure and estimated mean F_ST to be 0.032. Q_ST values for several floral traits including corolla tube length and width, colour, and nectar volume were large and significantly greater than mean F_ST. We performed multivariate comparisons of neutral loci to genetic correlations within and between populations and found a strong signal for divergent selection, suggesting that specific combinations of floral display and reward traits may be the targets of selection. Our results show little support for historical subspecies categories, yet floral traits are more diverged than expected due to drift alone. Non‐neutral divergence for multivariate quantitative traits suggests that selection by pollinators is maintaining a correlation between display and reward traits.     

The role of selection and historical factors in driving population differentiation along an elevational gradient in an island bird

Adaptation to local environmental conditions and the range dynamics of populations can influence evolutionary divergence along environmental gradients. Thus, it is important to investigate patterns of both phenotypic and genetic variations among populations to reveal the respective roles of these two types of factors in driving population differentiation. Here, we test for evidence of phenotypic and genetic structure across populations of a passerine bird (Zosterops borbonicus) distributed along a steep elevational gradient on the island of Réunion. Using 11 microsatellite loci screened in 401 individuals from 18 localities distributed along the gradient, we found that genetic differentiation occurred at two spatial levels: (i) between two main population groups corresponding to highland and lowland areas, respectively, and (ii) within each of these two groups. In contrast, several morphological traits varied gradually along the gradient. Comparison of neutral genetic differentiation (F_ST) and phenotypic differentiation (P_ST) showed that P_ST largely exceeds F_ST at several morphological traits, which is consistent with a role for local adaptation in driving morphological divergence along the gradient. Overall, our results revealed an area of secondary contact midway up the gradient between two major, cryptic, population groups likely diverged in allopatry. Remarkably, local adaptation has shaped phenotypic differentiation irrespective of population history, resulting in different patterns of variation along the elevational gradient. Our findings underscore the importance of understanding both historical and selective factors when trying to explain variation along environmental gradients.     

Selection on life‐history traits and genetic population divergence in rotifers

A combination of founder effects and local adaptation – the Monopolization hypothesis – has been proposed to reconcile the strong population differentiation of zooplankton dwelling in ponds and lakes and their high dispersal abilities. The role genetic drift plays in genetic differentiation of zooplankton is well documented, but the impact of natural selection has received less attention. Here, we compare differentiation in neutral genetic markers (F_ST) and in quantitative traits (Q_ST) in six natural populations of the rotifer Brachionus plicatilis to assess the importance of natural selection in explaining genetic differentiation of life‐history traits. Five life‐history traits were measured in four temperature × salinity combinations in common‐garden experiments. Population differentiation for neutral genetic markers – 11 microsatellite loci – was very high (F_ST = 0.482). Differentiation in life‐history traits was higher in traits related to sexual reproduction than in those related to asexual reproduction. Q_ST values for diapausing egg production (a trait related to sexual reproduction) were higher than their corresponding F_ST in some pairs of populations. Our results indicate the importance of divergent natural selection in these populations and suggest local adaptation to the unpredictability of B. plicatilis habitats.     

Low levels of quantitative and molecular genetic differentiation among natural populations of Medicago ciliaris Kroch. (Fabaceae) of different Tunisian eco-geographical origin

In this paper, we analyze the genetic variability in four Tunisian natural populations of Medicago ciliaris using 19 quantitative traits and six polymorphic microsatellite loci. We investigated the amplification transferability of 30 microsatellites developed in the model legume M. truncatula to M. ciliaris. Results revealed that about 56.66% of analyzed markers are valuable genetic markers for M. ciliaris. The most genetic diversity at quantitative traits and microsatellite loci was found to occur within populations (>80%). Low differentiations among populations at quantitative traits Q _ST  = 0.146 and molecular markers F _ST  = 0.18 were found. The majority of measured traits exhibited no significant difference in the level of Q _ST and F _ST . Furthermore, significant correlations established between these traits and eco-geographical factors suggested that natural selection should be invoked to explain the level of phenotypic divergence among populations rather than drift. There was no significant correlation between population differentiation at quantitative traits and molecular markers. Significant spatial genetic structure consistent with models of isolation by distance was detected within all studied populations. The site-of-origin environmental factors explain about 9.07% of total phenotypic genetic variation among populations. The eco-geographical factors that influence more the variation of measured traits among populations are the soil texture and altitude. Nevertheless, there were no consistent pattern of associations between gene diversity (He) and environmental factors.     

Contrasting patterns of morphological and neutral genetic divergence among geographically proximate populations of sockeye salmon Oncorhynchus nerka in Lake Aleknagik,Alaska

Levels of differentiation in morphological traits (age at maturity, body length at age, egg mass and body depth) and spawning time were examined in sockeye salmon Oncorhynchus nerka from three geographically proximate but physically distinct creeks in Lake Aleknagik, Alaska. Happy Creek fish had significantly greater values for most measured morphological traits, and Eagle Creek fish spawned significantly later than fish in the other creeks. Phenotypic differentiation between creeks, measured as P_ST, was then compared with microsatellite marker differentiation between creeks, measured as F_ST. No correlations were apparent between P_ST and F_ST values, and P_ST values were generally significantly larger than zero (P_ST= 0·0018–0·31) whereas F_ST values were not (F_ST=?0·0004 to 0·0016). The insignificant pair‐wise F_ST values between creek samples indicated that gene flow occurs between creeks, assuming the creek populations have reached migration–drift equilibrium. However, the strong homing behaviour of sockeye salmon precludes a scenario in which fish from the three creeks constitute a single population that segregates by body size. Rather, significant phenotypic differentiation suggests that strong divergent selection occurs on the phenotypic traits despite the homogenizing effects of gene flow.