Host race formation in phytophagous insects can be an early stage of adaptive speciation. However, the evolution of phenotypic plasticity in host use is another possible outcome. Using a reciprocal transplant experiment we tested the hypothesis of local adaptation in the aphid Brevicoryne brassicae. Aphid genotypes derived from two sympatric host plants, Brassica oleracea and B. campestris, were assessed in order to measure the extent of phenotypic plasticity in morphological and life history traits in relation to the host plants. We obtained an index of phenotypic plasticity for each genotype. Morphological variation of aphids was summarized by principal components analysis. Significant effects of recipient host on morphological variation and life history traits (establishment, age at first reproduction, number of nymphs, and intrinsic growth rate) were detected. We did not detected genotype × host plant interaction; in general the genotypes developed better on B. campestris, independent of the host plant species from which they were collected. Therefore, there was no evidence to suggest local adaptation. Regarding plasticity, significant differences among genotypes in the index of plasticity were detected. Furthermore, significant selection on PC1 (general aphid body size) on B. campestris, and on PC1 and PC2 (body length relative to body size) on B. oleracea was detected. The elevation of the reaction norm of PC1 and the slope of the reaction norm for PC2 (i.e., plasticity) were under directional selection. Thus, host plant species constitute distinct selective environments for B. brassicae. Aphid genotypes expressed different phenotypes in response to the host plant with low or nil fitness costs. Phenotypic plasticity and gene flow limits natural selection for host specialization promoting the maintenance of genetic variation in host exploitation.
Citation: Ruiz-Montoya L, Núñez-Farfán J (2013) Testing Local Host Adaptation and Phenotypic Plasticity in a Herbivore When Alternative Related Host Plants Occur Sympatrically. PLoS ONE 8(11): e79070. https://doi.org/10.1371/journal.pone.0079070
Editor: Norman Johnson, University of Massachusetts, United States of America
Received: May 17, 2013; Accepted: September 26, 2013; Published: November 12, 2013
Copyright: © 2013 Ruiz-Montoya, Núñez-Farfán. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: Financial support was provided by CONACYT grants (31543-B and 990039-DO) and by ECOSUR to LRM. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Host race formation or local host adaptation in phytophagous insects is considered an early stage of adaptive speciation in sympatry –. However, local adaptation to a host plant might be limited by the herbivore’s genetic diversity, gene flow and/or phenotypic plasticity , . Absence of genetic variation for host use (i.e., traits related to it) impedes selection to fit the best phenotype in each host. In turn, phenotypic plasticity implies that alternative phenotypes are produced by genotypes due to their environmental sensitivity , .
Phytophagous insects commonly interact with different host species at a local scale. Even at this scale, local adaptation to different host plants species is an expected outcome if selection in each host species is strong enough to prevent the homogenizing effects of gene flow. However, under soft selection (the probability of breeding after migration, cf. ), the plastic genotypes could be favoured over the local specialist genotypes –. Genotypes may respond differentially to each host by producing different phenotypes, those with the most favourable phenotype in each host persisting while others go extinct , . Although a single genotype might produce the best phenotype in all available hosts, plasticity may also induce a deviation far from the best local phenotype, reducing relative fitness of plastic genotype. Therefore, even though plasticity can potentially offer the best solution to herbivores in terms of phenotypic optima in different environments, costs and limits of plasticity may render plasticity suboptimal  or even maladaptive. The benefits of plasticity arise from phenotype-environment matching across different environments. Thus, adaptive genetic differentiation and adaptive phenotypic plasticity are two evolutionary paths to maximize fitness in response to environmental heterogeneity, and these responses are not mutually exclusive , .
To test the importance of phenotypic plasticity versus local adaptation in the interaction between a herbivorous insect and different host plants we studied the cabbage aphid, Brevicoryne brassicae Linnaeus (Homoptera: Aphididae), which feeds on host plants of Brassicacea family . In the state of Chiapas, south-eastern Mexico, this aphid reproduces parthenogenetically . Two common host plants of B. brassicae in this region are Brassica campestris L. and Brassica oleracea L., the former a superior host in terms of aphid fecundity and size , . Although differences in the probability of successful of establishment have been detected for aphids derived from different field hosts, suggesting local adaptation ; selection on the aphids reproductive span in each host is similar for aphids from both origins . Overall, the genetics of the aphid populations in the highlands of Chiapas revealed a moderate genetic structure (Fst = 0.22), however, very little genetic structure was explained by the host plant (Fst = 0.03) . Therefore, in this study, we tested the hypothesis of local adaptation in B. brassicae and assessed how the magnitude of phenotypic plasticity in morphological and life history traits is affected by selection. We analysed morphological and life history traits of clones (i.e., genotypes) derived from 28 aphid females collected and reared on both Brassica campestris and B. oleracea hosts.
Materials and Methods
Biological System of Study
Brevicoryne brassicae, an aphid species of palaearctic origin, is widely distributed and closely associated with host plant species of the Brassicaceae family . This aphid reproduces both sexually and asexually in cold regions of the world and completes its life cycle without host alternation. In regions with mild winters, like Chiapas, Mexico, reproduction is only by parthenogenesis .
B. brassicae and its cultivated host species B. oleracea (green cabbage) were probably simultaneously introduced to Mexico ca. 100 years ago . In the Chiapas highlands, the weed B. campestris and the cultivated B. oleracea var. capitata (green cabbage) are the main host plants of B. brassicae and both plant species commonly occur in sympatry. Green cabbage is cultivated from the end of autumn (November) to the end of winter (February). B. campestris is an annual weed that grows nearby from cultivated fields of B. oleracea. After harvest, B. campestris is tolerated and hence invades the fallow fields. In summer, small populations of B. brassicae can be found on discarded plants of B. oleracea during harvest or on late-emerging B. campestris plants. These host species present different environments for B. brassicae as B. campestris has a higher content of glucosinolates and higher leaf and stem trichome density than B. oleracea . In addition, B. campestris has a higher content of leaf nitrogen than B. oleracea .
Plants of B. campestris were grown from seeds collected in a field in the town of Teopisca (92°28′25″ W; 16°38′19″ N; 1800 m; annual average temperature, 17.25°C; annual average precipitation, 700 mm) (permission granted by Manuel Girón Intzín). B. oleracea seeds were obtained from a local farmer. Plants of both species were individually grown in pots in the greenhouse. Three seeds were sown in pots (20 d×30 h cm) with sterilized soil, and randomly assigned to benches in the greenhouse. One plant per pot was used to produce aphid clones or for performance tests. Each plant was covered with a mesh to exclude aphids and natural enemies . A total of 240 plants per species were produced. Plants were kept at ambient temperature and photoperiod (13±4°C and 12 h daylight), without fertilization, and watered every other day.
Life History Characters
A single parthenogenetic B. brassicae female was collected from each of twenty B. campestris plants and B. oleracea plants in Teopisca, making a total of 40 females. Each female and their offspring produced after one month were considered a genotype. From each genotype (clone) six nymphs were individually placed on B. campestris plants and six on the same number of B. oleracea plants. Each plant was covered with mesh. Each nymph was observed daily to determine if it successfully established itself on a host plant (i.e., ability to reproduce), as well as record its age at first reproduction, and the number of nymphs laid during a 15 days period of reproductive life (reproduction) (Table S1). A previous study demonstrates that most reproduction occurs during the 13 days period subsequent to onset of reproduction . Morphological and life history traits were not obtained for unsuccessful females and sample size was reduced for these traits. Seven genotypes collected on B. campestris (7/20) and five from B. oleracea (5/20) failed to establish successfully on one or both hosts.
We calculated the intrinsic growth rate per individual aphid according to the formula rm = 0.738 [(lnMd)/d], where d is age at first reproduction and Md is the number of descendants during a period of 15 days .
At the end of the experiment, each female was collected and kept in alcohol (70% solution). Then it was placed in a 10% KOH solution for 40 min, rinsed every half-hour during 8 h with distilled water, and cleared in a solution of chloral hydrate-phenol (1∶1) for at least 24 h (. Finally, specimens were placed on a slide; a drop of Berlesse medium was added and a cover glass was applied. Slides were oven-dried at 40°C for two weeks. Aphids were measured using a stereomicroscope (Stemi V6, Carl Zeiss). The length of body (BL), antennal segment III (ASIII), and hind tibia (HT) segment were measured (Table S1). These characters were chosen because they are involved in the recognition and use of the host plant , or associated host plant quality , . Variation in morphology was summarized by principal components analysis (PC).
Genetic differentiation between host species.
Differences among genotypes in relation to the host plant species were assessed using a mixed model analysis of variance (Anova) of age at first reproduction, number of nymphs, intrinsic growth rate, and on the scores of two principal components (PC1 and PC2). The variables age at first reproduction and number of nymphs were ln transformed to approximate experimental error to a normal distribution. The source host (i.e., origin) and recipient host (i.e., where grown) were declared as fixed effects, while genotype as a random effect nested within the source host. The host plant was declared as a fixed factor because the study was aimed to test local adaptation to these two particular host species known to elicit aphid variability in morphological and life history characteristics and occur in the same cultivated fields. Interactions between explanatory variables were included , . Establishment was analysed using a nominal logistic fit model. The response variable was recorded as successful (1) or unsuccessful (0) establishment as a function of the same independent variables described above. Significance of the effects was tested by Likelihood-ratio Chi-square tests . All analyses were carried out using the statistical software JMP 5.1.2. .
The interaction term source × recipient host in both the Anovas and the nominal logistic regression was of primary interest because it characterises genetic variation in the responses to recipient host. Differences among genotypes indicate genetic variation for the traits. Differences between recipient hosts suggest that these are different environments that affect performance of B. brassicae. The interaction genotype (nested in source host) × recipient host can be interpreted as evidence for local adaptation , if the effects are in the expected direction.
Phenotypic plasticity in morphological and life history traits of B. brassicae was analyzed by obtaining an index of plasticity for each genotype following Valladares et al. , calculating all pair-wise differences between individuals within a clone between environments (i.e., the two hosts). The index of plasticity was obtained for each character, for each genotype. This index is particularly relevant in this system because each aphid was allocated to one individual plant, representing a potentially additional environmental source of phenotypic variation within genotypes. The differences represent phenotypic plasticity for a given trait among individuals with the same genotype but that developed on different plants. A nested Anova of the Valladeres’ index of plasticity of each morphological (PC1 and PC2) and life history trait (age at first reproduction, number of nymphs laid, and intrinsic growth rate) in relation to the host of origin, and genotype nested within the host of origin, was performed to determine potential differences in plasticity. A significant host of origin effect indicates differences in the average index of plasticity between aphids collected on each plant, and significant genotype effect indicates differences in plasticity among genotypes.
Natural selection on morphological plasticity.
The reaction norm of each genotype for PC1 and PC2 was characterized by its elevation and slope , , . We regressed the average values of PC1 and PC2 of each genotype against the environments (host plants), declaring host as a continuous variable to obtain the slope and elevation of the reaction norm for each morphological trait per genotype. Slope and elevation represent the magnitude of phenotypic change induced by the hosts and the average trait value, respectively .
We tested if the slope and elevation of the reaction norm is related to the average relative fitness of genotypes by means of a multiple linear regression model , , . Relative intrinsic growth rate of each aphid was our estimator of fitness. Under this framework, a positive relationship between fitness and elevation is interpreted as the presence of selection for the maximum average body size across environments, whereas a positive relationship between the slope and fitness, suggests selection on body size plasticity; a negative relationship implies constraint plasticity . Regression analysis of the reaction norm components in relation to fitness has the potential problem of co-linearity . We explored these relationships by correlation analysis of trait plasticity and average trait values of genotypes within hosts , as well as randomization analysis of selection gradients as suggested by Roff .
The presence of plasticity costs was assessed by regression analysis of residual fitness (after regressing fitness against trait mean value) as a function of plasticity of the genotypes; a negative correlation would be indicative of costs . Fitness was the mean intrinsic growth rate across hosts, and the slope of the reaction norm was the magnitude of plasticity. This analysis was conducted for PC1 and PC2 separately.
In order to determine if phenotypic plasticity results from selection acting independently in each host, we estimated the selection gradient on PC1 and PC2 in each host plant species . We estimated the selection gradient by multiple regression analysis using standardized scores of PC1 and PC2 for each individual. Individual’s relative fitness was estimated as the intrinsic growth rate divided by the population mean intrinsic growth rate, thus . All analyses were performed using the JMP 5.1.2. statistical software .
Morphological and Life History Variation
Aphids’ morphological variation was summarized by PC1 and PC2, accounting 86.5% of the variance. PC1 accounted for 67% of variation (general body size), while 18.9% of variation (shape, size of appendages in relation to body size) was attributable to PC2 (Table 1). The general body size (PC1) and body length (PC2) had greater values in B. campestris than in B. oleracea (Figure 1 A, B).
Successful establishment of aphids was ca. 60% in B. oleracea and 80% in B. campestris, independent of host of origin (Fig. 1C). Genotypes differ in their probability to establish successfully (nested within origin; Likelihood ratio test, χ2 = 63.39, d.f. = 35, P = 0.0023) and interacted with the recipient host (Likelihood ratio test χ2 = 53.266, d.f. = 35, P = 0.024).
A significant recipient host effect was detected for all life history traits (Table 2). On average B. brassicae females initiated reproduction 1.5 days earlier and laid more nymphs on B. campestris than on B. oleracea (Fig. 1B, D). Similarly, the number of nymphs and intrinsic growth were higher in B. campestris than in B. oleracea (Fig. 1 E, F). In all cases, the effect of the origin host, genotype, and interactions (Fig. 2) were not significant (Table 2).
Points are the average value of 4–5 individuals of the same genotype. Genotypes were collected from B. campestris (solid lines) and B. oleracea (dashed lines).
Phenotypic Plasticity of Morphological and Life History Traits
The origin host did not affect the phenotypic plasticity index of B. brassicae genotypes (Table 3). In contrast, significant differences between aphid genotypes regarding the index of plasticity for all characters were detected (Table 3).
Selection of Morphological Traits
In each recipient, host positive directional selection on PC1 was detected (βB. campestris = 0.087; (βB. oleracea = 0.138). Aphids with larger body size (PC1) achieve more fitness; an increase in body size (PC1) by one standard deviation is expected to result in 8.7% more fitness than mean individual fitness in B. campestris (R2 = 0.055) and 13.8% in B. oleraceae (R2 = 0.152) (Table 4). Selection on PC2 on both recipient hosts was not significant, however it was marginally significant on B. oleracea (Table 4). Average mean relative fitness was higher on B. campestris (Fig. 1F).
Selection on Phenotypic Plasticity
The regression analysis of relative fitness on the reaction norm of PC1 was significant for the elevation (mean trait value; P = 0.052) but not for its slope (trait plasticity; P = 0.22). In contrast, positive directional selection on the slope of the reaction norm of PC2 was marginally significant but not for its elevation (Table 5). The bootstrap analyses show a potential distribution of selection gradients on plasticity from −0.029 to 0.181 for PC1, and from 0.028 up to 0.195 for PC2 (plasticity) (Table 5). The correlation of fitness residuals with the degree of plasticity was not significant for PC1 (R2 = 0.02, ß = 0.003, P = 0.48) or PC2 (R2 = 0.083, ß = 0.009, P = 0.14).
Correlation analysis of average trait value (mean trait) and trait plasticity (slope of reaction norm) differed between recipient hosts. This correlation was positive for PC1 and PC2 in B. oleracea but negative in B. campestris (Fig. 3).
For the most part, aphids grown on B. campestris attained, in general, larger body sizes (PC1) and relative body length (PC2) than those grown on B. oleracea, independent of genotype or host plant of origin. Both B. brassicae host plant species constitute distinct selective environments for B. brassicae, however the plastic response seems to involve low or nil fitness costs and plasticity in PC2 is positively selected (see ). Phenotypic plasticity and gene flow limits natural selection for host specialization promoting the maintenance of genetic variation in host use.
In the Chiapas highlands populations of B. brassicae are not locally adapted to host plants species, B. campestris and B. oleraceae, because local herbivore populations did not outperform the foreign ones. Aphids of both origins performed better when grown on B. campestris. Aphids morphological variation are determined by the particular host plant (B. oleracea and B. campestris) either in the field , or under experimental greenhouse conditions . In general, aphids grown on B. campestris are relatively thicker with long appendages, while aphids grown on B. oleracea are slightly flattened with short appendages. These phenotypes are elicited and predictable in these two Brassica species independent of the aphids’ genotype or their host plant of provenance. The difference in nitrogen content found between B. oleracea and B. campestris has been implicated as a source of phenotypic variation in B. brassicae . However a greenhouse experiment, in which the host plants were supplemented with soil nitrogen, failed to detect differences between hosts with regard to the elicited aphids’ morphological phenotype, but positively affected the number of nymphs laid . Plants’ phosphorous, potassium and glucosinolate content, which varies between hosts – has been shown to affect aphids’ development –. Thus it would be important to test if these play a role in the plastic morphology of B. brassicae.
Aphid genotypes of B. brassicae differed in their plasticity, but the Valladares’ index of plasticity was similar for aphids derived from B. campestris or B. oleracea. Since each individual female was established on a single plant, the plasticity index of a given genotype encompasses a fraction of the phenotypic variation elicited within a host plant species. Therefore, B. brassicae is not only highly sensitive to inter-specific but also to intra-specific differences among host plants. Can the distribution of phenotypic plasticity be partitioned in intra- and inter-specific components? Finding an answer to this question requires experimentally controlling the genetic variation of hosts, such that each aphid genotype develops on plants with known genotypes or the use of lines selected for low and high glucosinolates content . Thus, within-host variation precludes the detection of a genotype×host interaction.
Local adaptation to a host plant can be inferred from G×E interaction, with genotypes performing better in one host but not in the other . As mentioned, high interplant variation could be one of the causes, but a more significant one, is the high gene flow that might prevent genetic differentiation among hosts , at least for adaptive loci. Although B. campestris is the best host for B. brassicae, the high local abundance of the cultivated B. oleracea is a constant source, albeit less suitable, for aphids to colonize thus retarding/limiting the evolution of host local adaptation.
The host plant species plants provide distinct selective environments for B. brassicae and the optimal phenotype is different in each. The average phenotypic response occurs through plasticity at low fitness cost. van Tienderen  showed that in a coarse-grained environment consisting of two types of habitats, a population under soft selection is expected to evolve towards a compromise between the reaction that would be optimal within each habitat and a cost-free reaction norm, then predicted field populations composed of specialists, generalist, or intermediates or in a transitory state. This scenario may be represented by populations of B. brassicae in the Chiapas highlands where 20 genotypes reached the highest intrinsic growth rate in B. campestris; six genotypes performed better in B. oleracea, and six were generalists or canalized genotypes (i.e., equal intrinsic growth rate on both hosts). The specialists can use both hosts at nil or low cost of phenotypic plasticity. The relationship between fitness and plasticity in size (PC1) and relative body length (PC2) was positive and marginally significant. Plastic genotypes increased their fitness ca. 9.3%. We found a high correlation between the mean trait value and trait plasticity (PC1, PC2), but distinct between host plants (cf. Auld ). Trait value and its plasticity were positively correlated in B. oleracea but negatively in B. campestris, suggesting a biased estimate of the possible costs of plasticity. In both plants, more plastic genotypes had the most extreme phenotype, although in the opposite direction . Since larger general size renders higher fitness in B. campestris, plasticity covariate negatively with trait value and hence with fitness, resulting in a weak relationship between fitness and trait plasticity. The lack of a relationship between residual fitness and trait plasticity implies absence and/or nil cost of plasticity , , . Thus it is possible that selection has already removed genotypes of B. brassicae that incurred high costs of plasticity  or that plasticity in B. brassicae represent “noisy plasticity” . It appears that selection in each host is counterbalanced by plasticity.
We detected selection on aphid size (PC1) in B. campestris and B. oleracea indicating that the average phenotype is suboptimal in each host, while the relative body length is suboptimal in B. oleracea but not in B. campestris . This result reinforces the idea that these host species represent distinct selective environments for B. brassicae, as has been reported in previous studies , . Larger aphids attain higher fitness in both hosts but the average size is higher in B. campestris and thus the highest fitness is observed in this host, suggesting potential for host specialization of B. brassicae to B. campestris.
Host specialization of phytophagous insects is a relatively common phenomenon , , and may be a prelude of adaptive speciation . Disruptive selection has been considered as the main selection mode driving speciation under sympatric conditions–. We failed to detect whether this mode of selection, due to its relationship with fitness, affects size. Directional selection between hosts may promote differentiation if the selection differs in intensity . This proved to be true for B. brassicae, but perhaps selection intensity has not been strong enough to produce stronger genetic and phenotypic divergence , producing local adaptation to the host under sympatric condition . An additional constraint for evolution of local adaptation of B. brassicae in the Chiapas highlands is reproduction by parthenogenesis. Under this reproductive system the emergence of reproductive barriers necessary for adaptive speciation is precluded. Furthermore, as mentioned previously, there is a constant gene flow, via colonization, between host-associated populations , such that when coupled with plasticity, reduces the genetic differentiation .
In this study, phenotypic plasticity seems a sufficient for preventing the erosion of genetic variation by natural selection , , , delaying or impeding herbivore’s host race formation. Higher genotype×environment in relation to host use and strong selection within hosts could promote host race formation and incipient sympatric speciation, as has been documented in other aphid species, Acyrtosiphum pisum ,  Aphis fabae  and Myzus persicae , , although in the case of B. brassicae it seems to be a limited evolutionary path.
We are grateful to Manuel Girón Intzín and Guadalupe Andraca Gómez for their help in both the field and greenhouse work. We thank to C. D. Schlichting and an anonymous reviewer for the criticisms and suggestions. Thanks to R. Tapia López for logistic support and Julian Flavell for grammatical revision.
Conceived and designed the experiments: LRM JNF. Performed the experiments: LRM. Analyzed the data: LRM JNF. Contributed reagents/materials/analysis tools: LRM. Wrote the paper: LRM JNF.
- 1. Diehl S, Bush G (1984) An evolutionary and applied perspective of insect biotypes. Annual Review of Entomolgy 29: 471–504.
- 2. Mitter C, Farrell B, Futuyma DJ (1991) Phylogenetic studies of insect-plant interactions: Insights into the genesis of diversity. Trends in Ecology & Evolution 6: 290–293.
- 3. Fry JD (1996) The volution of host specialization: Are trade-offs overrated? The American Naturalist 148: S84–S107.
- 4. Tilmon K, Wood T, Pesek J (1998) Genetic variation in performance traits and potential for host shifts in Enchenopa Treehoppers (Homoptera: Membracidae). Annals of the Entomological Society of America 91: 397–403.
- 5. Schoonhoven LM (1999) Insects and plants: two worlds come together. Entomologia Experimentalis et Applicata 91: 1–6.
- 6. Loxdale HD, Lushai G, Harvey JA (2011) The evolutionary improbability of “generalism” in nature, with special reference to insects. Biological Journal of the Linnean Society 103: 1–18.
- 7. Sultan SE, Spencer HG (2002) Metapopulation structure favors plasticity over local adaptation. The American Naturalist 160: 271–283.
- 8. DeWitt TJ, Scheiner SM (2004) Phenotypic variation from single genotypes, a primer. In: DeWitt T, Scheiner S., editors. Phenotypic plasticity, functional and conceptual approaches. Oxford University Press, New York. 1–9.
- 9. Schlichting CD (1986) Evolution of phenotypiec plasticity in plants. Annual Review of Ecology and Systematics 17: 667–693.
- 10. Christiansen F (1975) Hard and soft selection in a subdivided population. The American Naturalist 109: 11–16.
- 11. Via S, Lande R (1985) Genotype-environment interaction and the evolution of phenotypic plasticity. Evolution 39: 505–522.
- 12. Bazzaz F, Sultan SE (1987) Ecological variation and the maintenance of plant diversity. In: Urbanska K, editor. Differentiation Patterns in Higher Plants. Academic Press, Orlando. 69–93.
- 13. van Tienderen PH (1997) Generalists, specialists, and the evolution of phenotypic plasticity in sympatric populations of distinct species. Evolution 51: 1372–1380.
- 14. Rausher MD (1984) Tradeoffs in performance on different hosts: Evidence from within- and between-site variation in the beetle Deloyala guttata. Evolution 38: 582–595.
- 15. Via S (1991) Specialized host plant performance of pea aphid clones is not altered by experience. Ecology 72: 1420–1427.
- 16. DeWitt TJ, Sih A, Wilson DS (1998) Costs and limits of phenotypic plasticity. Trends in Ecology & Evolution 13: 77–81.
- 17. Baythavong B, Stanton M (2010) Characterizing selection on phenotypic plasticity in response to natural environmental heterogeneity. Evolution 64: 2904–29020.
- 18. Núñez-Farfán J, Schlichting CD (2005) Natural selection in Potentilla glandulosa revisited. Evolutionary Ecology Research 7: 105–119.
- 19. Blackman R, Eastop V (2000) Aphids on the world’s crops. An identification and information guide. 2nd ed. Wiley, London.
- 20. Ruiz-Montoya L, Núñez-Farfán J (2009) Natural selection and maternal effects in life history traits of Brevicoryne brassicae (Homoptera: Aphididae) on two sympatric closely related hosts. Florida Entomologist 92: 635–644.
- 21. Ruiz-Montoya L, Núñez-Farfán J, Domínguez CA (2005) Changes in morphological traits of the cabbage aphid (Brevicoryne brassicae) associated with the use of different host plants. Ecological Research 20: 591–598.
- 22. Leal-Aguilar K, Ruiz-Montoya L, Perales H, Morales H (2008) Phenotypic plasticity of Brevicoryne brassicae in responses to nutritional quality of two related host plants. Ecological Entomology 33: 735–741.
- 23. Ruiz-Montoya L, Núñez-Farfán J (2004) Local adaptation of Brevicoryne brassicae (Homoptera: Aphididae) to host species at Chiapas, México. In: Simon J-C, Dedryver C-A, Rispe C, Hullé M, editors. Aphids in a New Millennium. Paris: Institut National de la Recherche Agronomique. 261–266.
- 24. Ruiz-Montoya L, Núñez-Farfán J, Vargas J (2003) Host-associated genetic structure of Mexican populations of the cabbage aphid Brevicoryne brassicae L. (Homoptera: Aphididae). Heredity 91: 415–421.
- 25. Rollins C (1993) The cruciferae of continental North America. Stanford University Press, Palo Alto.
- 26. Wayatt J, White P (1977) Simple estimation of intrinsic increase rates for aphids and tetranychid mites. Journal of Applied Ecology 14: 757–766.
- 27. Heie O (1987) Morphological structures and adaptations. In: Minks A, Harrewijn P, editors. World crop pests, aphids their biology, natural enemies and control. Elsevier, Elsevier. 393–400.
- 28. Ayres MP, Thomas DL (1990) Alternative formulations of the mixed-model anova applied to quantitative. Evolution 44: 221–226.
- 29. Fry JD (1992) The mixed-model analysis of variance applied to quantitative genetics: Biological meaning of the parameters. Evolution 46: 540–550.
- 30. Quinn G, Keough M (2002) Experimental design and data analysis for biologists. Cambridge University Press, Cambridge.
- 31. SAS I (2003) JMP. Statistics and graphics guide version 5.1.01.2. North Carolina, Care: SAS Institute.
- 32. Valladares F, Sanchez-Gomez D, Zavala MA (2006) Quantitative estimation of phenotypic plasticity: bridging the gap between the evolutionary concept and its ecological applications. Journal of Ecology 94: 1103–1116.
- 33. De Jong G (1990) Genetics of reaction norms. Journal of Evolutionary Biology 3: 447–468.
- 34. Gavrilets S, Scheiner SM (1993) The genetics of phenotypic. V. Evolution of reaction norm shape. Journal of Evolutionary Biology 48: 31–48.
- 35. Lande R, Arnold SJ (1983) The measurement of selection on correlated characters. Evolution 37: 1210–1226.
- 36. Auld JR, Agrawal AA, Relyea RA (2010) Re-evaluating the costs and limits of adaptive phenotypic plasticity. Proceedings of the Royal Society of London B-Biological Sciences 277: 503–511.
- 37. Roff DA (2011) Measuring the cost of plasticity: a problem of statistical non-independence. Proceedings of the Royal Society of London B-Biological Sciences 278: 2724–2725.
- 38. van Tienderen PH (1991) Evolution of generalists and specialist in spatially heterogeneous environments. Evolution 45: 1317–1331.
- 39. Ruiz-Montoya L, Leal-Aguilar K, Núñez-Farfán J (2012) Evolutionary importance of phenotypic plasticity in aphid-host interactions: Insight from a study of Mexican populations of Brevicoryne brassicae. In: Jenkins O, editor. Advances in Zoology Research. New York: Nova Science Publisher, Inc., Vol. 4: 179–197.
- 40. Uddin MM, Ulrichs C, Tokuhisa JG, Mewis I (2009) Impact of glucosinolate structure on the performance of the crucifer pest Phaedon cochleariae (F.). Journal of Applied Botany and Food Quality 82: 108–113.
- 41. Kos M, Kabouw P, Noordam R, Hendriks K, Vet LEM, et al. (2011) Prey-mediated effects of glucosinolates on aphid predators. Ecological Entomology 36: 377–388.
- 42. Wang H, Wu J, Sun S, Liu B, Cheng F, et al. (2011) Glucosinolate biosynthetic genes in Brassica rapa. Gene 487: 135–142.
- 43. Annan B, Ampong-Nyarko K, Tingey W, Schaefers G (1997) Interactions of fertilizer, cultivar selection, and infestation by cowpea aphid (Aphididae) on growth and yield of cowpeas. International Journal of Pest Management 43: 307–312.
- 44. Miyasaka SC, Hansen JD, McDonald TG, Fukumoto GK (2007) Effects of nitrogen and potassium in kikuyu grass on feeding by yellow sugarcane aphid. Crop Protection 26: 511–517.
- 45. Cole RA (1997) The relative importance of glucosinolates and amino acids to the development of two aphid pests Brevicoryne brassicae and Myzus persicae on wild and cultivated brassica species. Entomologia Experimentalis et Applicata 85: 121–133
- 46. Stowe KA (1998) Experimental evolution of resistance in Brassica rapa: Correlated response of tolerance in lines selected for glucosinolate content. Evolution 52: 703–712.
- 47. Kawecki TJ, Ebert D (2004) Conceptual issues in local adaptation. Ecology Letters 7: 1225–1241.
- 48. Van Buskirk J, Steiner U (2009) The fitness costs of developmental canalization and plasticity. Journal of evolutionary biology 22: 852–860.
- 49. Via S (1994) The evolution of phenotypic plasticity: What do we really know? In: Real L, editor. Ecological genetics. Princeton, N.J.: Princeton University Press. 35–57.
- 50. Farrell BD, Sequeira AS (2004) Evolutionary rates in the adaptive radiaton of beetles on plants. Evolution 58: 1984–2001.
- 51. Cook LG, Rowell DM (2007) Genetic diversity, host-specificity and unusual phylogeography of a cryptic, host-associated species complex of gall-inducing scale insects. Ecological Entomology 32: 506–515.
- 52. Caillaud MC, Via S (2000) Specialized feeding behavior influences both ecological specialization and assortative mating in sympatric host races of Pea Aphids. The American Naturalist 156: 606–621.
- 53. Mayr E (1966) Evolution and the diversity of life, selected essays. Cambridge, MA: Harvard University Press.
- 54. Maynard Smith J (1966) Sympatric speciation. The American Naturalist 100: 637–650.
- 55. Via S (2001) Sympatric speciation in animals: the ugly duckling grows up. Trends in Ecology & Evolution 16: 381–390.
- 56. Dhuyvetter H, Hendrick F, Gaublomme E, Desender K (2007) Differentiation between two salt marsh beetle ecotypes: evidence for ongoing speciation. Evolution 61: 184–193.
- 57. Weis AE, Abrahamson W, Andersen MC (1992) Variable selection on Eurosta’s gall size, I: The extent and nature of variation in phenotypic selection. Evolution 46: 1674–1697.
- 58. Fitzpatrick BM (2012) Underappreciated consequences of phenotypic plasticity for ecological speciation. International Journal of Ecology 2012: 1–12.
- 59. Spitze K, Sadler TD (1996) Evolution of a generalist genotype: multivariate analysis of the adaptiveness of phenotypic plasticity. The American Naturalist 148: S108–S123.
- 60. Gorur G, Cecilia L, Mackenzie A (2005) Phenotypic plasticity in host-plant specialisation in Aphis fabae. Ecological Entomology 30: 657–664.
- 61. Borja Peppe F, Lomonaco C (2003) Phenotypic plasticity of Myzus persicae (Hemíptera: Aphididae ) raised on Brassica oleracea L. var acephal (kale) and Raphanus sativus L. (radish). Genetics and Molecular Biology 26: 189–194.
- 62. Cardoza YJ, Wang SF, Reidy-Crofts J, Edwards OR (2006) Phloem alkaloid tolerance allows feeding on resistant Lupinus angustifolius by the aphid Myzus persicae. Journal of Chemical Ecology 32: 1965–1976.