Introductions of non-native species can result in strong selective challenges for native populations. The strength of selection in this case is determined by the size of the environmental shift, which imposes a fitness cost on the population proportional to the squared distance between the population mean phenotype and the position of the new optimum 12. If the optimum moves far enough, the fitness cost will be sufficiently high to reduce the intrinsic rate of increase of the population to <1. Unless the population can rapidly advance toward the new optimum phenotype, it will not persist 3. Two processes can facilitate persistence of populations challenged with a rapidly changing environment: adaptive phenotypic plasticity and genetic adaptation.

Adaptive phenotypic plasticity allows individuals within a population to accommodate a changing environment 45 by facilitating rapid movement to a new fitness optimum. This movement occurs through changes in the mean value of a trait and/or changes in the genetic and phenotypic variance/covariance structures. In the extreme case, plastic changes in the mean value of a trait are able to completely move a population to a new fitness optimum and no genetic adaptation is required 5. In cases where a plastic change in the mean is not sufficient to shift a population to a new optimum it can allow a population to persist until sufficient adaptive genetic changes occur 67. An incomplete shift in the population mean towards a new selective optimum that facilitates population persistence is called the "Baldwin effect" 6 and the resulting reduction in the intensity of selection is referred to as the "Bogert effect" 8 or adaptive buffering 9. Plastic changes in the (co)variance matrix may result in increased levels of expressed genetic variance (i.e., variance that is context-dependent and arises only in response to specific environmental cues) and changes in covariances between traits that increase the response to selection 10111213.

Populations may also adapt genetically to new environmental conditions when there is no pre-existing adaptive phenotypic plasticity or plasticity is insufficient to completely shift a population to a new phenotypic optimum. The rate of genetic adaptation toward a new optimum is determined by a number of factors, including the amount of additive genetic variation present for the traits under selection 14, the rate at which mutation produces new adaptive variation 1516, and genetic correlations among characters 16171819.

A common source of rapid environmental change arises from the introduction of novel predator species. In a notable example, non-native fishes have been widely introduced into naturally fishless alpine lakes throughout the world and have had profound effects on native zooplankton species, including Daphnia. Daphnia have a long history as a model system to study the consequences of introduced fish predators 20212223. Daphnia adapt to introduced fish through changes in traits related to detection avoidance, including alterations in patterns of diel vertical migration (DVM) 2425 and reduced body size 212326. Daphnia also display significant adaptive phenotypic plasticity in response to chemical cues produced by fish that can facilitate persistence during changes in selection regime. Plastic changes that reduce pigmentation 27 and body size 2829303132 in Daphnia decrease the ability of fish to detect their prey resulting in higher survivorship, while plastic increases in fecundity 282930 result in higher intrinsic rates of population increase.

Daphnia melanica (identified as Daphnia middendorfiana in previously published studies, e.g., 333435, but recently classified as D. melanica based on molecular analyses [M. Pfrender, unpublished data]) populations located in alpine lakes throughout the Sierra Nevada in eastern California, USA provide a unique opportunity to study the effects of introduced predators on naive populations. These alpine lakes have been the subjects of extensive ecological study 333435 in part because the history of fish introductions is well documented. In lakes where D. melanica and fish co-occur, D. melanica have smaller body sizes and reproduce earlier relative to those in lakes without fish 36. These differences were attributed to rapid adaptive evolution. However, because Daphnia are often highly plastic in response to chemical cues from fish, the differences in morphology and life-histories observed previously may not be solely due to underlying genetic alteration. Differences in morphology and life-history could be entirely due to phenotypic plasticity or a combination of plastic and genetic modification.

To determine the relative contributions of adaptive plasticity and genetic adaptation during rapid evolution in response to introduced fish we conducted common-garden experiments on clonally reproducing females of D. melanica populations collected from four lakes in the Sierra Nevada with contrasting fish stocking histories. Two lakes were never stocked and remain in their natural fishless condition and two lakes have contained introduced fish populations during the last several decades. We measured morphological and life-history traits of clonally reproducing females from each population in the presence and absence of chemical cues from fish (i.e., fish kairomone). Because D. melanica can be maintained in a state of constant clonal reproduction in the lab, it is straightforward to utilize standard quantitative genetic techniques to estimate the contribution of genetic and plastic phenotypic effects underlying adaptive traits. Our chief working assumption in this experiment is that the phenotypic states of fishless populations are representative of the ancestral phenotypic states of populations that currently contain fish. Given our assumption of equality between currently fishless populations and ancestral states of fish populations is true, our study design allowed us to determine the degree of morphological and life-history differentiation due to selection by fish predation and quantify the contribution of phenotypic plasticity in determining adaptive responses to the introduction of fish.

Rapid evolution is an important component of the success of invading species 37 and the response of organisms in invaded communities 9 because it ameliorates the selective cost imposed by a shift in the phenotypic optimum and enhances the probability of long-term population persistence. Similarly, adaptive plasticity may be an important component of rapid evolution as it can allow short-term population persistence following changes in the selective environment 68 that in turn provides time for evolutionary mechanisms to operate. However, disentangling actual cases of rapid evolution from purely plastic changes in response to a changing environment can be problematic because adaptive phenotypic plasticity is a common feature in many organisms 38.

In this study we examined the rapid changes of Daphnia morphology and life histories in response to a single abrupt change in the environment. Specifically, we investigated the relative role of genetic and plastic phenotypic changes in moving populations exposed to a novel predator toward a new fitness optimum. Our results show that reductions in egg number and body size of D. melanica genotypes from the Sierra Nevada, in response to introduced fish, are largely adaptive evolutionary responses and not due entirely to adaptive phenotypic plasticity. We do find evidence for adaptive plasticity, in the form of increases in clutch size and reductions in body size, in these populations that could facilitate short-term persistence and subsequent rapid evolution. We caution, however, that our interpretation of these results is predicated on the assumption that our measured phenotypes of currently fishless populations are representative of the ancestral phenotypes of populations that currently co-exist with fish.

Our results suggest that naive D. melanica populations in the Sierra Nevada may initially respond to fish introductions through adaptive phenotypic plasticity brought about by chemical cues from fish, which facilitates movement towards the new phenotypic optimum. First, plastic reductions in body size make D. melanica less visible to fish and constitute evidence for a "Baldwin/Bogert effect" 68. Fish are highly effective size-selective predators and their efficiency is primarily linked to prey visibility 26394041. Thus, D. melanica that are less visible have a fitness advantage (via increased survivorship) through movement towards the new phenotypic optimum and the resulting reduction in selection intensity due to decreased predator efficiency. Concomitant with a decrease in body size, D. melanica also show adaptive phenotypic increases in fecundity that could lead to higher intrinsic rates of population increase. Although the rate at which fish remove individuals from these Daphnia populations is unknown, our observed increase in clutch size of approximately one is quite significant. Estimates of D. melanica population sizes in the Sierra Nevada are on the order of hundreds of millions to billions [R. Knapp, unpublished data], thus, an increase of one individual at first reproduction might substantially offset any losses due to predation. An interesting aspect of our findings is that our naive Daphnia populations, those without any history of fish exposure, are responsive to chemicals produced by fish. This observation suggests that D. melanica may be pre-adapted to fish predation, and that the genetic machinery responsible for adaptive phenotypic plasticity in response to fish kairomone is ancestral in this species.

Although we find evidence for adaptive phenotypic plasticity that would facilitate short-term population persistence in the face of novel predation, it is not sufficient to explain the difference in body-size and egg number between populations that are historically fishless and those that co-occur with fish. For example, a comparison of the average body size in the Evelyn Lake population (1.34 mm in the kairomone(-) and 1.24 mm in the kairomone(+) treatments) with the average in the fishless populations (2.00 mm in the kairomone(-) and 1.95 mm in the kairomone(+) treatments) shows that the mean phenotype in Evelyn Lake has diverged by 9.4 phenotypic standard deviations. The change in body size attributable to plasticity in fishless populations is approximately one standard deviation. In other words, the change in body size due to plasticity accounts for only about 11% of the total difference observed between Evelyn and fishless populations. Thus, the phenotypic differences observed in our study appear largely due to changes in the underlying genetic components controlling phenotype.

Our observation that the body-size response in Evelyn Lake was much higher than that in Puppet Lake could arise for three reasons. First, the difference in body size could simply reflect the different amounts of time each population was exposed to fish predation (Puppet Lake – 53 years; Evelyn Lake – 91 years). Second, Daphnia populations may have experienced differing levels of fish predation resulting in varying selection intensities, with the selection intensity in Evelyn Lake substantially higher. Finally, given our observation that expressed levels of genetic variance for body size increased approximately 6-fold in response to fish kairomone in one fishless population (Frog Lake) but not in the other (Source Lake) our fish populations may have differed in the initial levels of standing genetic variation, with the Evelyn Lake population harboring more standing genetic variation than Puppet Lake.

Our observation that Evelyn Lake was the only population to display a significant evolutionary reduction in egg number is likely due to the ability of D. melanica to deposit large amounts of melanin in the carapace. Melanin production in the carapace would initially "blind" selection to changes in egg number. Thus, an evolutionary response in egg number should occur only after reductions in melanin deposition. Fish predation produces strong selection on melanin production in other Daphnia populations 42, and there is evidence for reduced melanin expression in Daphnia from our fish populations relative to fishless populations [M. Pfrender, unpublished data]. Therefore, the apparent delayed onset of selection on egg number could be due to initial selection on melanin production and subsequent selection on egg number.

Traditional views of character evolution typically involve trade-offs among traits that can limit the adaptive potential of a population 43. However, several selection experiments involving Daphnia suggest adaptive evolutionary changes in one trait are not necessarily associated with concomitant maladaptive changes in others 444546. We observe a similar result here, where evolutionary and plastic changes in body size and fecundity occur in the absence of changes in the timing of maturity and reproduction. Our results, and those of other researchers that imply the absence of a trade-off, could be attributed to assay conditions in which food is not a limiting resource 47. Daphnia morphology and life-history can display food concentration-dependent reactions to the presence of fish kairomone 48.

We investigated the relative contributions of selection and adaptive phenotypic plasticity to the rapid evolution of morphology and life histories in response to an introduced predator. We conclude that adaptive plasticity could facilitate short-term population persistence through "Baldwin/Bogert effects", but that long-term persistence was achieved through subsequent genetic adaptation. Further investigation into other traits that may have also undergone rapid change in selective regime as a consequence of fish introductions, such as pigmentation and DVM behavior, examined under differing kairomone and food conditions, will provide a more detailed view of the traits and processes involved in the overall evolution of the Daphnia/fish predator-prey system in the Sierra Nevada.

Numerous studies have examined the contributions of plasticity and selection to rapid adaptation in non-native species following their introduction into a novel environment (e.g., 49). In contrast, few studies have examined the phenotypic and evolutionary response of native species to introduced species that pose strong novel selective challenges. Thus, this study and a growing body of others investigating the response of native communities to introduced species should enhance our understanding of the conditions necessary for rapid adaptive evolution and the relationship between rapid evolution and population persistence 9.

LCL designed and conducted the experiment, performed statistical analyses, and drafted the manuscript. JWB conducted the experiment. RAK provided background information on Daphnia populations, assisted with fieldwork in the Sierra Nevada, and helped draft the manuscript. MEP conceived of the study, conducted fieldwork in the Sierra Nevada to collect the experimental populations, and contributed to analysis and manuscript preparation. All authors read and approved the final manuscript.