The Journal of Heredity 2001:92(1)
© 2001 The American Genetic Association 92:30-37
Age-Specific Fitness Components in Hybrid Females of Drosophila pseudoobscura and D. persimilis
From the Department of Genetics, University of Georgia, Athens, GA 30602-7223.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Most models of hybridization assume that hybrids are less fit than their parental taxa. In contrast, some researchers have explored the possibility that hybrid individuals may actually have higher fitness and so play an important role in the generation of new species or adaptations. By estimating age-specific fitness components, we can determine not only how hybrid fitness differs from parental taxa, but also whether the fitness of hybrids relative to parental taxa changes with age. Here we describe an analysis of age-specific fitness traits in two species of Drosophila, D. pseudoobscura and D. persimilis, and their F1 hybrids. At early ages, hybrid females lay as many eggs as parental individuals, on average, but produce far fewer offspring. By late ages, in contrast, parental taxa show a steep decline in production not seen in hybrids, such that hybrids produce more offspring, on average, than parental taxa. Furthermore, egg-adult survival in hybrids is negatively correlated with egg density, whereas these traits are only weakly correlated in parental taxa. The results are limited somewhat by the fact that we analyze only two strains, and that these may be partially inbred. Nonetheless, the results are certainly illustrative, pointing out not only that at least some hybrid individuals may be as fit or fitter than parental taxa, but also that the difference between hybrids and parental taxa varies with age.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
For much of this century natural hybridization has been recognized as a major contributor to the evolution of plant and animal species complexes (Anderson 1949; Anderson and Hubricht 1938; Dowling and DeMarais 1993; Grant and Grant 1992; Lotsy 1916; Moritz 1983; Shaw et al. 1985; Stebbins and Daly 1961). However, studies of natural hybridization and, in particular, investigations of the effect of natural selection and gene flow on the evolution of animal hybrid "zones" have been mainly limited to inferences derived from the genetic structure itself (e.g., Heywood 1986; Marchant et al. 1988; but see Rand and Harrison 1989). This has often been due to the experimental limitations of the species under investigation. Most hybrid zone analyses have thus examined hybridization between animal taxa for which experimental manipulations are difficult (see Barton and Hewitt 1985, Hewitt 1988 for species lists). Where experimentation has been possible, it has resulted in a more detailed description and understanding of the factors determining the outcome of gene exchange (e.g., Shaw et al. 1980).
However, to fully examine the potential evolutionary importance of hybridization as a force for generating novel species or adaptations, we need to describe how natural selection acts on hybrid populations over evolutionary time. This is necessary because all models that predict the evolutionary trajectory of hybrid populations (natural and experimental) assume a primary role for natural selection. For example, the model most commonly applied (the tension zone model—Barton and Hewitt 1985, Key 1968) assumes that hybrids are uniformly less fit than their parental taxa in all habitats. The more recently proposed "mosaic model" also has hybrid unfitness as a central tenet, at least when applied to the outcome of reinforcement (Harrison 1986; Howard 1982, 1986). In contrast, the bounded hybrid superiority model (Moore 1977) and the evolutionary novelty model (Arnold 1997) assume that some hybrid genotypes can be more fit than their parents in certain environments. The role that natural selection might assume in the evolution of hybrid populations has, however, rarely been directly tested. Rather the form (i.e., positive or negative) of selection has been incorporated as a major assumption in the various models; an assumption that is not tested by the outcomes predicted by the various models (Moore and Price 1993). Thus one of the two major factors (the other being gene flow; Barton and Hewitt 1985) believed to be causal in the evolution of hybrid populations and taxa has not been rigorously examined.
We can easily determine relative fitness in hybrids and parental species by comparing demographic traits among these populations (Fisher 1930). In addition, we can use estimates of age-specific fecundity and survival to measure how selection pressures might shape future changes in life-history strategy (Caswell 1989; Charlesworth 1994). Although a few analyses have estimated the fitness of specific hybrid genotypes (e.g., Burke et al. 1998a,b; Hotz et al. 1999; Spaak and Hoekstra 1995; Weider and Wolf 1991), almost none have considered the effect of interspecific hybridization on age-specific components of fitness (but see Dudycha and Tessier 1999). The most direct way to determine how the fitness of a hybrid differs from that of its parental strains is to estimate age-specific survival and fecundity in hybrid and parental strains. Previous studies have shown that, at least under certain conditions and for certain traits, hybrids can perform as well as or better than their parental species (see Arnold 1997 for a review). Our aim here is to determine whether the relative fitness components of female hybrids and parental species varies across age classes. This information may help us to determine under which demographic and environmental regimes hybrids are most likely to play a role in speciation. For example, in declining populations, the relative frequency of older individuals can increase. If older individuals play a disproportionate role in hybrid fitness, then hybridization could be an effective way of "rescuing" a population in sustained decline.
Such an approach may also help in the search for genes that influence hybridization and speciation (e.g., Orr 1987; Palopoli and Wu 1994). Recent studies with Drosophila melanogaster suggest that there are genes whose effects on fitness are confined to specific ages (Hughes and Charlesworth 1994; Pletcher et al. 1998; Promislow et al. 1996). By incorporating an age-specific component into studies of hybridization and fitness, we may be able to identify genes that affect fitness in hybrids, but whose effects are confined to specific ages.
To explore these ideas, we will measure age-specific fitness components in hybrid females of two species of fruit fly, D. pseudoobscura and D. persimilis. D. pseudoobscura is found throughout western and southwestern North America, from Texas to British Columbia. D. persimilis has a smaller range, occurring solely on the west coast of North America and within the range of D. pseudoobscura. These two species provide an ideal model system to study the effects of hybridization on age-specific fitness components. They are easily cultured in the laboratory and have relatively similar life-history strategies. D. persimilis and D. pseudoobscura will readily mate with one another in the laboratory to produce viable F1 offspring (Anderson et al. 1972). Hybrid females are fertile or semisterile, while males are sterile (Dobzhansky 1951; Mampell 1941; Orr 1987). Under experimental conditions, males of both species will court heterospecific females, though a majority of females demonstrate assortative mating behavior (Dobzhansky 1951; Noor 1995). Although hybrids are thought to be rare in nature [Dobzhansky (1973) estimated that less than 1 in 6000 flies collected in the wild had hybridized, and this may have occurred in transport from the wild to the laboratory], there is strong molecular evidence of regular gene flow between these two species for at least some loci (Powell 1983; Wang et al. 1997). In spite of large, if not complete, reproductive isolation between these two species, they still provide a powerful laboratory model to study the interaction of age structure and genetic structure in the formation of hybrids.
Extensive work has explored the genetic basis of mating behavior in hybrids and parental lines of these two species (Noor 1995, 1996; Noor and Aquadro 1998; Noor and Coyne 1996), the relative fitness of hybrid genotypes (e.g., Hutter and Rand 1995; Orr 1987, 1989), and age-specific fitness traits within species (Anderson and Watanabe 1997). In this study we will compare age-specific female fecundity and mortality, and larval viability in cohorts of D. pseudoobscura, D. persimilis, and their two reciprocal F1 hybrids. The analysis is somewhat limited by its use of only one strain each for the two parental species. Nonetheless, we hope that by integrating genetic and demographic approaches, we might begin to understand in general how demography influences patterns of speciation (e.g., Marzluff and Dial 1991), and in particular, whether certain genotypes are more likely to contribute to successful hybrid populations.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Strains of two species of fruit flies, D. pseudoobscura and D. persimilis, were obtained from the laboratory of Dr. Wyatt Anderson (Department of Genetics, University of Georgia). The D. pseudoobscura line was collected in Mesa Verde, Colorado, in 1995, and the strain of D. persimilis was collected in Mather, California, in 1995. Prior to starting the experiment, stocks were maintained in half-pint bottles on standard molasses-agar-cornmeal medium at 19°C on a 12 h light/12 h dark cycle. During the course of the experiment we kept all flies in 8 dram glass vials on standard medium at 19°C.
Hybridization
To obtain F1 hybrid flies we carried out reciprocal interspecific crosses between virgin males and virgin females of D. pseudoobscura and D. persimilis. Individual pairs were placed in 8 dram glass vials on standard molasses-agar-cornmeal medium with live bakers yeast added and allowed to mate for 48 h. For each of the four possible pairwise crosses (D. pseudoobscura females x D. persimilis males, D. persimilis females x D. pseudoobscura males, D. persimilis females x D. persimilis males, D. pseudoobscura females x D. pseudoobscura males) we set up 25 separate vials. Each pair was then transferred to a new vial and allowed to lay eggs for 6 days. Flies were then removed and vials of fertilized eggs were kept at 19°C until the F1 generation began to emerge, approximately 19 days later. Eclosing flies were collected as virgins and placed in single-sex vials on standard medium, with approximately 20 flies per vial.
To confirm that pure species strains and hybrids were not contaminated, we analyzed salivary polytene chromosomes for eight larvae from each of the four crosses. In all cases, chromosome squashes confirmed that the crosses were as expected. Hereafter, we refer to offspring of D. pseudoobscura females and D. persimilis males as "pseud-pers," and offspring of D. persimilis females and D. pseudoobscura males as "pers-pseud."
Fecundity and Fertility
We estimated fertility and fecundity in 50 females from each of the four strains at four different ages. Before placing females in vials, we placed one hundred 2- to 3- day-old D. persimilis males and 100 D. pseudoobscura males individually in vials containing standard medium seeded with live yeast and allowed them to acclimate for 24 h. We then added a single hybrid or pure-strain female to each vial. Females from the F1 crosses were placed with a pure-strain male that was the same species as the female parent (i.e., pers-pseud females with D. persimilis males and pseud-pers females with D. pseudoobscura males).
Flies were transferred to new vials every 5 days, at which time we determined if the female and male were still alive. Females were given virgin males approximately every 2 weeks to ensure a sufficient supply of sperm.
We carried out the first analysis at day 6, which ensures that all flies had reached reproductive maturity. At four ages—6, 20, 36, and 52 days—we placed single females in fresh vials and allowed them to lay eggs for 24 h. Flies were then transferred to new vials. Vials set up at ages 6, 20, and 36 days were maintained until all offspring had eclosed, and the adult flies were counted. From these counts we also obtained egg-to-adult survival rates for each female at each of three ages. For the vials set up at 52 days, we counted eggs, but did not determine egg-adult survival. The experiment was terminated when the flies were 67 days old.
Within each cohort, egg counts and offspring number were not normally distributed among flies. To test for differences in egg counts and offspring number among lines, we used the nonparametric Mann–Whitney U test (Sokal and Rohlf 1995).
Within-Strain Variability
It is generally assumed that hybrids are less fit than their pure-strain progenitors. However, some genotypes can demonstrate elevated fitness relative to their parents (see Arnold 1997 for a review).
Accordingly, to compare variation in fecundity and larval viability among lines, we tested for variation in these traits among all fecund individuals and also among the top 10% (5 flies) in each line at each age. We used a randomization approach (Manly 1997) to mitigate the problems of small sample size, with 5000 randomizations for each pairwise comparison.
Egg-Adult Viability
In addition to our analysis of egg counts and offspring number, we also compared line means for egg-adult viability, defined as the proportion of eggs in a vial that survive to eclose as adults. This was done for eggs laid at 6, 20, and 36 days, but not for eggs laid at 52 days.
Viability is a binomially distributed trait—when few eggs are laid, the error variance for the estimate of viability is very high. To account for this, in our comparisons of means for egg-adult viability, we weighted all samples by the square-root of egg number. We denote egg-adult viability, the ratio of flies emerging to eggs laid by the ith fly, as vi. The weighted mean viability, 
, is given by
 | (1) |
Density Effects on Egg-Adult Viability
In laboratory fruit fly populations, egg viability typically decreases as larval density increases (Sang 1949), due to limited resources and increased concentration of metabolic waste products. In this experiment we determined the relationship between viability and density for each of the four lines at each of three ages.
In a regression of egg viability on egg density, the error variance is expected to be binomially distributed (eggs either live or die). Standard linear regression models assume that error variance is normally distributed. Thus to test for a relationship between egg density and egg-adult viability, we used a general linear model for logistic regression (Sokal and Rohlf 1995) with a binomial error variance in GLIM (Numerical Algorithms Group, Oxford). The model is given by
 | (2) |
is equal to log{p/(1 - p)}, where p is the proportion of eggs surviving to eclosion. The model treats emerging flies as the dependent variable and egg number as the independent variable, with the error term dependent on the original egg number. The output from this model gives a slope 
and intercept ß from which we can calculate the probability of egg survival p as a function of egg number x where
 | (3) |
Adult Survival Rate
Although the focus of the experiment was on fecundity measurements, we kept data on survival for individual flies through the end of the experiment (67 days). To test for differences among lines in adult survival, we used Kaplan-Meier and proportional hazards models in JMP® (SAS Institute, Cary, NC). For flies that survived to the end of the experiment, we included these as censored individuals in the statistical tests (Lee 1992).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Egg Counts
On average, 81% of flies laid at least one egg at any one time (Figure 1). After excluding flies that laid no eggs at all, we found that at all four ages both hybrid lines generally laid as many eggs as the pure strains, with the exception of pers-pseud at 6 days, which laid fewer eggs than the other three lines (Figure 2A). At 36 days pure D. persimilis flies laid significantly fewer eggs than pseud-pers and D. pseudoobscura. By 52 days both hybrid lines laid more eggs on average than either of the pure lines, though only pseud- pers was significantly greater than the parental individuals.
|
|
A similar pattern emerged when we considered the mean egg counts among the five most fecund flies at each time point (Figure 2B). The most fecund hybrid flies were similar to the D. pseudoobscura flies throughout the experiment. However, the most fecund flies from the pure D. persimilis strain laid significantly fewer eggs than the hybrids at three of the four time points. Similarly the number of D. persimilis flies laying no eggs increased to more than 50% by the end of the experiment, compared with less than 10% for D. pseudoobscura flies and approximately 20% for the hybrid lines (Figure 1).
To correct for multiple comparisons, we used a sequential Bonferroni correction (Holm 1979). In the case of egg counts, there were no longer any significant differences between lines. However, the comparison is extremely conservative. Within each set of comparisons there were 24 pairwise tests for egg counts. Thus the revised = 0.05/24 = 0.0021. In the case of egg counts, none of the comparisons reached this level of significance. However, for mean egg production (Figure 2A), one-third of all comparisons were significant at P < .05 before the correction. It is highly unlikely that one would observe these many significant differences by chance. Thus although we present Bonferroni corrections for all figures, we also point out those comparisons that were statistically significant before correction.
Offspring Number
If we consider the average number of flies emerging at each time point, the pattern is quite different from what we see for egg counts. In general, the pure strains produce more flies per vial than the hybrids, though at the latest age, as with egg counts, D. persimilis flies produce fewer flies than either hybrid strain (Figure 3).
|
D. pseudoobscura shows a slight decline and D. persimilis shows a strong decline in offspring productivity over time. In contrast, pseud-pers productivity declined substantially from the first assay to the second and then increased by the third assay. Productivity in pers-pseud increased steadily over the course of the experiment.
When we compared the five most productive flies in each line, we found that D. pseudoobscura flies consistently produced more flies than all other lines. However, during each of the first two assays, one hybrid was not significantly lower than D. pseudoobscura (pseud-pers at day 6, pers-pseud at day 20). Overall D. pseudoobscura flies had higher productivity than the other three lines, and the two hybrids were not significantly different from D. persimilis flies.
Although we found that the best among the oldest hybrids (day 52) laid more eggs than the parental strains, the experiment was not continued long enough to assay for survival of those eggs to adulthood.
Egg-to-Adult Viability
Hybrid females frequently produced as many eggs as the pure parental strains. However, the probability of eggs surviving to adulthood was markedly lower in the hybrids than in the parental lines for females at age 6 and 20 days. This is consistent with previous studies that have found hybrid females to be semisterile (Mampell 1941; Orr 1987). By 36 days, the difference between hybrids and D. persimilis in egg viability had disappeared, with all lines having viability of approximately 50% (Figure 4). D. pseudoobscura egg viability remained high.
|
Viability and Density
We compared the relationship between viability and density for each of the four lines at three time points, for 12 comparisons in all (Figure 5). In five of the six hybrid assays, viability declined significantly as density increased. In only one of the six parental strain assays did viability decrease significantly with density. Thus egg viability appears to be much more adversely affected by density in hybrids than in pure strains.
|
Adult Survival
Survival rates of adults did not differ among any of the lines (Kaplan–Meier log rank test,
2 = 4.37, df = 3, P = .22; proportional hazard 2 = 2.78, df = 3, P = 0.43). By the end of the experiment, 42% of the 199 initial individuals were still alive (28 of 50 pers-pseud, 17 of 50 pseud-pers, 23 of 50 D. pseudoobscura, and 15 of 49 D. persimilis). Given the relatively low mortality and small sample sizes, our statistical power was very low, so actual differences in survival rates may not have been detected. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The standard paradigm in studies of interspecific hybrids has been that hybrids are less fit than their pure parental types or strains (Arnold 1997). In contrast, it has been suggested that within a population, a few hybrid individuals or genotypes may have fitness that equals or even exceeds that of their parents, and that these hybrids may play a critical role in the origin of novel species (Anderson and Stebbins 1954; Arnold 1997; Arnold and Hodges 1995).
The study we describe here demonstrates that, in the case of D. pseudoobscura x D. persimilis hybrids, there are indeed a few individuals in which offspring number matches or exceeds that of one or both parental types. Under natural, competitive conditions, although rare, the few hybrids that are fitter than any of the parents may contribute disproportionately to the opportunity for gene flow. Of greater interest, we find that the differences in fitness components between hybrids and parental strains vary with age. At early ages, fitness components are generally greater in the parental strains. By late ages, however, this difference is reduced, particularly between D. persimilis parents and both hybrids.
Our conclusions are somewhat limited by the fact that we have studied one strain each of D. pseudoobscura and D. persimilis, both of which may be somewhat inbred. Fitness components in the parental strains may have declined due to expression of late-acting deleterious recessive alleles, which are more likely to go to fixation than early acting alleles with deleterious effects (Medawar 1952). If such alleles exist and are fixed in either of the parental strains, then among hybrids, effects of late-acting recessive deleterious alleles in one species could be "rescued" by wild-type alleles in the other species, and so ameliorate the decline in fitness components.
In addition, we collected data for egg viability from a relatively narrow range of ages. At the oldest age (day 52), hybrids had egg numbers equal to or greater than the parental species, suggesting that fitness may have been greater, but we do not have viability estimates for this last age. Future experiments would clearly benefit from extending the analysis to very late ages.
Despite these caveats, the approach we present here demonstrates the potential value of adding age structure to studies of hybridization. Until now, few studies have incorporated age-structure in studies of hybrid fitness (but see Dudycha and Tessier 1999). If the finding that hybrid individuals exhibit relatively greater fitness components than pure species at older ages turns out to be generally true among hybrids, it may provide a hint at the demographic regime in which we might be most likely to see hybrids emerge successfully, or at least to contribute genetically to future generations. In rapidly growing populations, where fitness contributions are biased toward young individuals, genetic contributions from pure species are likely to predominate over hybrids. In contrast, in declining populations, selection will favor delayed reproduction and the age structure will shift toward predominantly older individuals (Charlesworth 1994). In this setting, overall fitness differences between hybrids and parentals may be drastically reduced, and hybrid genotypes may contribute substantially to later generations. The effect of demographic structure on hybridization may be extended to overall environmental quality. For example, in a study of hybridization in D. persimilis and D. pseudoobscura, Van Valen (1963) found that at 25°C, stable hybrid swarms between the two species formed, whereas at 16°C they did not.
What might account for this age-related increase in the fitness components of hybrids relative to parental strains? One possibility may have to do with how selection has shaped reproductive strategies. In the rapidly increasing populations that are typical of insects such as Drosophila, in the wild (Andrewartha and Birch 1954) as well as the laboratory (e.g., Mueller and Ayala 1981), there is a premium on high investment in early reproduction (Clark 1987; Promislow and Tatar 1998; Rowe L and Houle D, personal communication). In the pure strains we see a relatively high investment in reproduction early, followed by a subsequent decline. As previous work has shown, costs of this early reproduction may be paid not only in terms of decreased late-age survival (e.g., Partridge 1987; Rose and Charlesworth 1980; Zwaan et al. 1995), but also in terms of reduced late-age fecundity. Of interest, we do not see an age-related decline in reproductive output in hybrids.
What might account for the apparent lack of a cost of reproduction in the hybrids? The parental strains had been kept in laboratory culture, in which early reproduction is strongly favored (Promislow and Tatar 1998), even if it reduces late-age fitness. In the hybrids we have disrupted the evolved genomic structure of the organism. One consequence of this may be that hybrids lose the ability to regulate the delicate timing of age-specific reproduction. They may no longer be able to deliberately put most of their reproductive effort into early age reproduction. As a consequence, reproductive output is more constant throughout the life span of hybrids (Figures 2 and 3). Hybrids would end up paying less of a cost for early reproduction than pure strains, and so have greater resources to allocate to reproduction late in life when compared with the pure parental strains. In the short term, under specific demographic conditions, this change in life-history strategy may favor hybrids, and could even account for the limited gene flow that has been observed between D. persimilis and D. pseudoobscura (Wang et al. 1997). As we mentioned previously, this result may also be due to the masking in hybrids of late-acting deleterious recessive alleles.
Genetic effects specific to individual parental species may also determine the success of hybrids. In our study, we found D. pseudoobscura outperformed D. persimilis in all traits and at all ages. Hybrids produced by D. persimilis females crossed with D. pseudoobscura males were less vigorous than the reciprocal hybrids, suggesting a strong effect of D. persimilis X-linked or mitochondrial genes on fitness. Orr (1987) observed that backcross males carrying a D. pseudoobscura X chromosome were almost always fertile, while backcross males carrying a D. persimilis X chromosome were almost always sterile, though there may also be a strong positive effect of the D. pseudoobscura mitochondrial genome on hybrid fitness (Hutter and Rand 1995). Similarly, Noor (1997) found that hybrid males with a D. persimilis X chromosome displayed much weaker courtship behavior than either of the pure species or the reciprocal hybrid. Noor (1997) also found strong epistatic effects among loci. The present study was not designed to detect such effects, though given Noor's result, as well as recent work on epistatic effects in other species (e.g., Burke et al. 1998a,b), future studies of fitness in Drosophila hybrids should explore this possibility in greater detail.
Egg Survival versus Egg Number
In hybrids, higher density leads to a decrease in egg-adult survival, while in pure strains there is no such effect of density (Figure 5). There are two possible explanations for this difference. First, at higher densities, competition among larvae is likely to be more intense. Larvae from hybrid mothers may be less able to cope with these stressful conditions than larvae from parental mothers, so the pattern is seen in hybrids but not parental strains. Second, to the extent that nongenetic factors influence the fitness of eggs (Mousseau and Fox 1998), there may be a trade- off between the number of eggs that a fly can lay at one time and the quality of those eggs, and this trade-off may be more evident in hybrids than in parental strains. If this trade-off leads to reduced hatching rate in hybrids, then the overall density of adults may not differ between hybrids and parental strains. Thus the effect of high density may be due not to competition among larvae, but rather to the inability of hybrid females to produce viable eggs when laying many of them.
Our interpretation of these results must be taken with the following two caveats. First, the statistic that we used related number of adults/number of eggs to the number of eggs. Thus the built-in autocorrelation would lead us to expect a negative correlation. However, we would not expect a difference in pattern between hybrid and parental taxa, so this explanation is unlikely. Second, in this experiment, variation among females in density was determined by the females rather than by the experimenter. Thus although viability decreased as density increased in hybrids, there may be no direct causal connection between the two traits. We cannot be sure whether the effect is due to the direct influence of egg density, or to some other factor that causes some females both to lay many eggs and to have relatively low egg viability.
Given that viability is not affected by egg number in parental strains, this result suggests that there is directional selection for increased fecundity. By contrast, in hybrids, since viability decreases with egg density, the maximum number of offspring will be produced at an intermediate number of eggs, leading to stabilizing selection on egg number. Thus the selective landscape on age-specific fitness components may be fundamentally different for hybrids than for pure strains, and new niches, at least with respect to life-history strategies, might be available to hybrids.
|
Acknowledgments |
|---|
We thank W. W. Anderson for providing the strains used in this experiment and for help with cytotyping the hybrids, and L. Pearse for assistance in the lab. We received helpful comments on a previous version of this manuscript from W. W. Anderson, A. Keyser, D. Hoyt, C. Spencer, and L. Yampolsky. This work was supported by a National Institute on Aging grant AG14027 (to D.E.L.P.) and a National Science Foundation grant DEB-9703853 (to M.L.A.).
|
Footnotes |
|---|
Address correspondence to Daniel E. L. Promislow at the address above or e-mail: promislo{at}arches.uga.edu.
Corresponding Editor: Stephen Schaeffer
Received March 3, 2000
Accepted August 31, 2000
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-
Anderson E, 1949. Introgressive hybridization. New York.: John Wiley & Sons.
Anderson E and Hubricht L, 1938. Hybridization in Tradescantia. III. The evidence for introgressive hybridization. Am J Bot 25:396–402.
Anderson E and Stebbins GL Jr, 1954. Hybridization as an evolutionary stimulus. Evolution 8:378–388.
Anderson WW, Dobzhansky T, and Pavlovsky O, 1972. A natural population of Drosophila transferred to a laboratory environment. Heredity 28:101–107.[Web of Science][Medline]
Anderson WW and Watanabe TK, 1997. A demographic approach to selection. Proc Natl Acad Sci USA 94:7742–7747.
Andrewartha HG and Birch LC, 1954. The distribution and abundance of animals. Chicago: University of Chicago Press.
Arnold ML, 1997. Natural hybridization and evolution. Oxford: Oxford University Press.
Arnold ML and Hodges SA, 1995. The fitness of hybrids—a response to Day and Schluter. Trends Ecol Evol 10:289.
Barton NH and Hewitt GM, 1985. Analysis of hybrid zones. Annu Rev Ecol Syst 16:113–148.[Web of Science]
Burke JM, Carney SE, and Arnold ML, 1998a. Hybrid fitness in the Louisiana irises: evidence from experimental analyses. Evolution 52:37–43.
Burke JM, Voss TJ, and Arnold ML, 1998b. Genetic interactions and natural selection in Louisiana iris hybrids. Evolution 52:1304–1310.
Caswell H, 1989. Matrix population models. Sunderland, MA: Sinauer.
Charlesworth B, 1994. Evolution in age-structured populations, 2nd ed. Cambridge: Cambridge University Press.
Clark AG, 1987. Senescence and the genetic correlation hang-up. Am Nat 129:932–940.
Dobzhansky T, 1951. Experiments on sexual isolation in Drosophila. X. Reproductive isolation between Drosophila pseudoobscura and Drosophila persimilis under natural and under laboratory conditions. Proc Natl Acad Sci USA 37:792–796.
Dobzhansky T, 1973. Is there gene exchange between Drosophila pseudoobscura and Drosophila persimilis in their natural habitats? Am Nat 107:312–314.[Web of Science]
Dowling TE and DeMarais BD, 1993. Evolutionary significance of introgressive hybridization in cyprinid fishes. Nature 362:444–446.
Dudycha JL and Tessier AJ, 1999. Natural genetic variation of life span, reproduction, and juvenile growth in Daphnia. Evolution 53:1744–1756.
Fisher RA, 1930. The genetical theory of selection. Oxford: Clarendon Press.
Grant PR and Grant BR, 1992. Hybridization of bird species. Science 256:193–197.
Harrison RG, 1986. Pattern and process in a narrow hybrid zone. Heredity 56:337–349.
Hewitt GM, 1988. Hybrid zones—natural laboratories for evolutionary studies. Trends Ecol Evol 3:158–167.
Heywood JS, 1986. Clinal variation associated with edaphic ecotones in hybrid populations of Gaillardia pulchella. Evolution 40:1132–1140.
Holm S, 1979. A simple sequential rejective multiple test procedure. Scand J Stat 6:65–70.
Hotz H, Semlitsch RD, Gutmann E, Guex GD, and Beerli P, 1999. Spontaneous heterosis in larval life-history traits of hemiclonal frog hybrids. Proc Natl Acad Sci USA 96:2171–2176.
Howard DJ, 1982. Speciation and coexistence in a group of closely related ground crickets. New Haven, CT: Yale University Press.
Howard DJ, 1986. Zone of overlap and hybridization between two ground cricket species. Evolution 40:34–43.
Hughes KA and Charlesworth B, 1994. A genetic analysis of senescence in Drosophila. Nature 367:64–66.[Medline]
Hutter CM and Rand DM, 1995. Competition between mitochondrial haplotypes in distinct nuclear genetic environments: Drosophila pseudoobscura vs. D. persimilis. Genetics 140:537–548.[Abstract]
Key KHL, 1968. The concept of stasipatric speciation. Syst Zool 17:14–22.
Lee ET, 1992. Statistical methods for survival data analysis, 2nd ed. New York: John Wiley & Sons.
Lotsy JP, 1916. Evolution by means of hybridization. The Hague, Netherlands: M. Nijhoff.
Mampell K, 1941. Female sterility in interracial hybrids of Drosophila pseudoobscura. Proc Natl Acad Sci 27:337–341.
Manly BFJ, 1997. Randomization, bootstrap and Monte Carlo methods in biology, 2nd ed. New York: Chapman & Hall.
Marchant AD, Arnold ML, and Wilkinson P, 1988. Gene flow across a chromosomal tension zone I. Relicts of ancient hybridization. Heredity 61:321–328.
Marzluff JM and Dial KP, 1991. Life history correlates of taxonomic diversity. Ecology 72:428–439.
Medawar PB, 1952. An unsolved problem in biology. London: H. K. Lewis.
Moore WS, 1977. An evaluation of narrow hybrid zones in vertebrates. Q Rev Biol 52:263–277.[Web of Science]
Moore WS and Price JT, 1993. Nature of selection in the northern flicker hybrid zone and its implications for speciation theory. In: Hybrid zones and the evolutionary process (Harrison RG, ed). Oxford: Oxford University Press; 196–225.
Moritz C, 1983. Parthenogenesis in the endemic Australian lizard Heteronotia binoei (Gekkonidae). Science 220:735–737.
Mousseau TA and Fox CW, 1998. Maternal effects as adaptations. Oxford: Oxford University Press.
Mueller LD and Ayala FJ, 1981. Trade-off between r-selection and K-selection in Drosophila populations. Proc Natl Acad Sci USA 78:1303–1305.
Noor MA, 1995. Speciation driven by natural selection in Drosophila. Nature 375:674–675.[Medline]
Noor MA and Coyne JA, 1996. Genetics of a difference in cuticular hydrocarbons between Drosophila pseudoobscura and D. persimilis. Genet Res 68:117–123.[Web of Science][Medline]
Noor MAF, 1996. Absence of species discrimination in Drosophila pseudoobscura and D. persimilis males. Anim Behav 52:1205–1210.
Noor MAF, 1997. Genetics of sexual isolation and courtship dysfunction in male hybrids of Drosophila pseudoobscura and Drosophila persimilis. Evolution 51:809– 815.
Noor MAF and Aquadro CF, 1998. Courtship songs of Drosophila pseudoobscura and D. persimilis: analysis of variation. Anim Behav 56:115–125.[Web of Science][Medline]
Orr HA, 1987. Genetics of male and female sterility in hybrids of Drosophila pseudoobscura and Drosophila persimilis. Genetics 116:555–563.
Orr HA, 1989. Localization of genes causing postzygotic isolation in two hybridizations involving Drosophila pseudoobscura. Heredity (Edinburgh) 63:231–237.
Palopoli MF and Wu CI, 1994. Genetics of hybrid male-sterility between Drosophila sibling species—a complex web of epistasis is revealed in interspecific studies. Genetics 138:329–341.[Abstract]
Partridge L, 1987. Is accelerated senescence a cost of reproduction? Funct Ecol 1:317–320.
Pletcher SD, Houle D, and Curtsinger JW, 1998. Age-specific properties of spontaneous mutations affecting mortality in Drosophila melanogaster. Genetics 148:287–303.
Powell JR, 1983. Interspecific cytoplasmic gene flow in the absence of nuclear gene flow: evidence from Drosophila. Proc Natl Acad Sci USA 80:492–495.
Promislow DEL and Tatar M, 1998. Mutation and senescence: where genetics and demography meet. Genetica 102/103:299–314.
Promislow DEL, Tatar M, Khazaeli A, and Curtsinger JW, 1996. Age-specific patterns of genetic variance in Drosophila melanogaster: I. Mortality. Genetics 143:839–848.[Abstract]
Rand DM and Harrison RG, 1989. Ecological genetics of a mosaic hybrid zone: mitochondrial, nuclear, and reproductive differentiation of crickets by soil type. Evolution 43:432–449.
Rose MR and Charlesworth B, 1980. A test of evolutionary theories of senescence. Nature 287:141–142.[Medline]
Sang JH, 1949. The ecological determinants of population growth in a Drosophila culture. III. Larval and pupal survival. Physiol Zool 22:183–202.
Shaw DD, Coates DJ, Arnold ML, and Wilkinson P, 1985. Temporal variation in the chromosomal structure of a hybrid zone and its relationship to karyotypic repatterning. Heredity 55:293–306.
Shaw DD, Moran C, and Wilkinson P, 1980. Chromosomal reorganization, geographic differentiation and the mechanism of speciation in the genus Caledia. In: Insect cytogenetics (Blackman RL, Hewitt GM, and Ashburner M, eds). Oxford, Blackwell Scientific Publications; 171–194.
Sokal RR and Rohlf FJ, 1995. Biometry, 3rd ed. New York: W. H. Freeman.
Spaak P and Hoekstra JR, 1995. Life history variation and the coexistence of a Daphnia hybrid with its parental species. Ecology 76:553–564.[Web of Science]
Stebbins GL and Daly K, 1961. Changes in the variation pattern of a hybrid population of Helianthus over an eight-year period. Evolution 15:60–71.
Van Valen L, 1963. Introgression in laboratory populations of Drosophila persimilis and D. pseudoobscura. Heredity 18:205–214.
Wang RL, Wakeley J, and Hey J, 1997. Gene flow and natural selection in the origin of Drosophila pseudoobscura and close relatives. Genetics 147:1091–1106.[Abstract]
Weider LJ and Wolf HG, 1991. Life history variation in a hybrid species complex of Daphnia. Oecologia 87:506–513.
Zwaan B, Bijlsma R, and Hoekstra RF, 1995. Direct selection on life-span in Drosophila melanogaster. Evolution 49:649–659.[Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||