ORIGINAL ARTICLE

Lorena Ruiz-Montoya Æ Juan Nunez-Farfan

Cesar A. Domınguez

Changes in morphological traits of the cabbage aphid (Brevicorynebrassicae) associated with the use of different host plants

Received: 9 August 2004 / Accepted: 11 March 2005 / Published online: 19 May 2005� The Ecological Society of Japan 2005

Abstract The genetic and morphological differentiationof insect populations in relation to the use of differenthost plants is an important phenomenon that predatesecological specialisation and speciation in sympatricconditions. In this study, we describe the morphologicalvariation of populations of Brevicoryne brassicae (Ho-moptera: Aphididae) associated with two host species,Brassica oleracea and Brassica campestris, which occursympatrically in the highlands of Chiapas, Mexico. Thestudy is aimed at obtaining evidence regarding pheno-typic differentiation induced by, or associated with, theuse of distinct but closely related host species. Sevenmorphological characters were measured in 696 winglessaphids collected from plants of the two host species atfour localities. Morphological variation was summarisedthrough principal components analysis (PCA). Sixty-two percent of morphological variation was explainedby the first two PCs. The first component (PC1) wasrelated to the general size of appendages, and PC2 wasinterpreted as the relationship between body size (bodyand leg size) and antenna length. Aphids growing on B.campestris were bigger than those collected from B. ol-eraceae. Significant differences between hosts were de-tected for PC1, whereas a significant effect of locality,host, and the interaction locality · host was detected forPC2. These results indicate that the average phenotypeof B. brassicae individuals inhabiting different host-plant

species differs as a consequence of the contrastingfeeding environments the host species provide.

Keywords Brevicoryne brassicae Æ Host specialisation ÆLocal adaptation Æ Phenotypic plasticity Æ Phenotypicvariation Æ Plant–insect interaction

Introduction

Geographically isolated populations may differ due tofounder effects and low migration rates, or becauseselection favours some genotypes over others in eachenvironment. If ecological specialisation is promoted byselection, then phenotypic and genetic differentiationamong populations may occur even at short distances(Schluter 1995; Nagel and Schluter 1998). Specialisationin the use of different resources within a habitat mayproduce intense selective pressures and promote localadaptation and sympatric speciation (Bush 1994;Schluter 2001; Via 2002).

The study of the interaction between plants andherbivorous insects has underlined the importance ofhost specialisation as a mechanism of speciation undersympatric conditions (Bush 1994; Schluter 2001;Berlocher and Feder 2002; Via 2001, 2002). Althoughmost studies have focused on changes in life-historytraits, few have documented morphological changes re-lated to the specialisation process (Nosil and Crespi2004). The use of different host plants may favour thedivergent evolution of traits related to feeding efficiency.It has been documented, for instance, that the averagebeak length of the soapberry bug Jadera haematoloma(Hemiptera) varies in accordance with the average seedsize of its host (Carroll and Boyd 1992).

Individuals of the same species that have developed indifferent environments, e.g. localities or host plants, maydiffer considerably in morphological characters of eco-logical relevance. Generally, these changes have a directrelationship to individual fitness (Roff 1992; Via and

L. Ruiz-MontoyaEl Colegio de la Frontera Sur,Carretera Panamericana y Periferico Sur s/n,San Cristobal de Las Casas, CP, 29290 Chiapas, Mexico

J. Nunez-Farfan (&) Æ C. A. DomınguezLaboratorio de Genetica Ecologica y Evolucion,Departamento de Ecologıa Evolutiva,Instituto de Ecologıa,Universidad Nacional Autonoma de Mexico,Circuito exterior s/n, Ciudad Universitaria,CP, 04510 Coyoacan, Distrito Federal, MexicoE-mail: farfan@servidor.unam.mx

Ecol Res (2005) 20: 591–598DOI 10.1007/s11284-005-0076-3

Shaw 1996; Nagel and Schluter 1998; Pappers et al.2001), suggesting that phenotypic evolution is driven bynatural selection. Phenotypic differences, however, maybe purely environmental in origin if genotypes are plasticand their reaction norms vary due to differences in theirsensitivity to the environment (Via 1987; Weisser andStadler 1994; Schlichting and Pigliucci 1998). Accord-ingly, the evolution of ecological specialisation versusgeneralism in phytophagous insects depends on thepresence of different host species occurring sympatri-cally, on the amount of gene flow, the frequency of eachenvironment, and the fitness differences between spe-cialists and generalists in a given environment (Moran1992; Sultan and Spencer 2002).

Aphids (Homoptera) are herbivorous insects withmonophagous and polyphagous species (i.e., feeding onfew or several host plant species, respectively; Fox andMorrow 1981; Blackman and Eastop 2000). Evenpolyphagous species show some level of specialisation toone particular host species in a given locality (Takada1979; Moran 1981; Weber 1985; Anstead et al. 2002).Aphids constitute an excellent system to study the con-sequences of different host usage on the patterns ofmorphological variation, and for the evolution of sym-patric specialisation. Morphological characters ofaphids are highly variable (Eastop 1985) and may beassociated with the species of host and with microcli-matic conditions (Blackman and Paterson 1986; Moran1986; Blackman 1987; Heie 1987; Damsteegt and Vo-egtlin 1990; Sokal et al. 1991; Lazzari and Voegtlin 1993;Ruiz-Montoya et al. 1994; Via and Shaw 1996; Dixon1998).

In this study we describe the patterns of morpho-logical variation of populations of Brevicoryne brassicaeL. (Homoptera: Aphididae) associated with two closelyrelated host species (Brassica oleracea L. and Brassicacampestris L.) within and among localities in the high-lands of Chiapas, Mexico.

Materials and methods

Study area

The study area is located in the highlands of Chiapas insouthern Mexico at altitudes between 2,000 and 2,500 m(Fig. 1). The climate is temperate sub-humid with amean annual rainfall of 1,100 mm and a mean annualtemperature of 16�C. The rainy season lasts from Mayto October. The landscape is severely fragmented withmany fields of agricultural crops surrounded by sec-ondary oak forests (Quercus spp.) in different succes-sional stages (Gonzalez-Espinosa et al. 1997).

The aphid species

B. brassicae L. is an oligophagous herbivore that feedsupon species of the plant family Brassicaceae and is

distributed in temperate regions of the world (Black-man and Eastop 2000). B. brassicae may display bothsexual and asexual phases on an individual plant(Blackman and Eastop 2000). In the highlands ofChiapas B. brassicae feeds mainly upon B. campestrisand B. oleracea, and probably has an anholocyclic lifecycle (only parthenogenetic reproduction), with wingedand wingless morphs reaching highest density in March(L. Ruiz-Montoya, personal observation). There is noevidence of sexual reproduction in the populations ofthis aphid in Chiapas (e.g., presence of oviparous fe-males, eggs or males; L. Ruiz-Montoya, personalobservation).

Previous studies indicated the presence of significantgenetic differentiation among populations of B. brassi-cae from different localities and between host plantswithin localities (Ruiz-Montoya et al. 2003). Also,reciprocal transference experiments indicate thatB. brassicae reproduce earlier and produce more nymphsin B. campestris than in B. oleracea (Ruiz-Montoya andNunez-Farfan 2004).

Host plants

B. oleracea (green cabbage) and B. campestris (syn.B. rapa, turnip; Rollins 1993) were introduced in Mexicoprobably 100 years ago (Rollins 1993), and both speciesare frequently used by B. brassicae as hosts. These twospecies commonly occur in sympatry in the highlands ofChiapas. Green cabbage has been annually cultivated inChiapas for at least the last five decades (Montoya-Gomez 1998). Turnip is a ruderal annual plant thatgrows near cultivated fields of green cabbage or inabandoned fields. Both hosts are available to aphidsmainly during the winter months (from October toFebruary). Small populations of B. brassicae can befound during summer on harvested plants of B. olera-ceae left in the field, or on late-emerged plants ofB. campestris.

B. campestris has a higher content of glucosinolates,and higher leaf and stem trichome density than B. oler-aceae (Rollins and Banerjee 1976; Rollins 1993). Thesetwo characters are considered plant defences againstinsects (Louda and Mole 1991; Agren and Schemske1994; Stowe 1998). In addition, B. campestris has ahigher content of nitrogen and lipids than B. oleraceae(L. Ruiz Montoya, unpublished data). Thus, B. oleraceaand B. campestris potentially constitute two differentenvironments for B. brassicae.

Sampling and measurements of aphids

B. brassicae individuals were collected at four localitiesin the highlands of Chiapas during February and March1999 (Fig. 1). In each locality, aphids were haphazardlycollected from 15–20 individual plants of each hostspecies. Aphids from each plant were kept in 70%

592

alcohol in separated labelled vials. From each plant, wemounted 1–2 wingless adults on slides until 30 aphidsper host plant per locality had been completed. We fo-cused on the wingless individuals because they are al-most sessile and spend at least three generations in agiven plant. Winged forms appear when the plant isnutritionally poor and they usually leave the plant.Thus, winged individuals in a given plant may havecome from different plants or even different host species.

Before measurements, specimens were placed in a10% KOH solution for 40 min, then rinsed every half-hour for 8 h with distilled water and then cleared in asolution of chloral hydrato-phenol (1:1) for at least 24 h(Blackman and Eastop 2000). Finally, specimens wereplaced in a small drop of Berlesse medium, and a coverglass was applied and the slide was oven-dried at 40�Cfor 2 weeks.

Aphids were measured using a stereomicroscope(Stemi V6, Carl Zeiss). The length of seven characterswas measured to the nearest mm: body (BL), antennalsegment III (AIII), base of antennal segment VI (BVI),terminal process of last antennal segment (TP), lastrostral segment (LRS), hind tibia (HTIB) and hindtarsal segment II (HTII) (Fig. 2). These characterswere chosen because they are involved in the recog-nition and use of the host plant (Heie 1987). Bodylength is frequently associated with reproductive ratesince smaller aphids reproduce more slowly than larger

ones (Dixon 1998). Hind tibia and hind tarsal arerelated to an aphid’s mobility and its capacity to reachthe host plant or to escape from natural enemies(Shaposhnikov 1985). Longer antennae are supposedto improve the ability of orientation (Shaposhnikov1985), and the last rostral segment is strongly corre-lated with the selection of leaf surfaces of host plants,such as the presence and density of trichomes (Moran1986; Heie 1987).

Statistical analyses

Variability of morphological characters of B. brassicaewas analysed using both univariate and multivariatemethods. In order to reduce the number of variables, aprincipal components analysis was carried out using thecorrelation matrix of the seven morphological traitsmeasured using the JMP statistical package (SASInstitute 1995). Because scores are linear combinationsof the original variables, they may be used as inte-grated descriptors of the general morphology of aphids.Scores from the first two PCs were analysed throughanalysis of variance. In these analyses, we declaredlocality as a random factor, while host was kept as afixed effect (Underwood 1997). The same approach wasemployed for the analysis of individual morphologicalcharacters.

Fig. 1 Location and environmentalcharacteristics of populations of B.brassicae in Chiapas (Mexico) (geographicposition, altitude above sea level, annualmean temperature and precipitation)

593

Results

Two principal components explained 62% of the varia-tion in the morphology of B. brassicae. The first com-ponent explained 50% of the variance and the second12% (Table 1). All characters had positive loadings onPC1. The highest loadings on this component wereassociated with variation in the size of the appendages.Body length and the two attributes related to the size ofthe legs had positive loadings on PC2. Attributesdescribing the variation in antenna and mouthparts hadnegative loadings on this component (Table 1).Accordingly, scores from PC1 were used as descriptorsof the general size of the appendages. Scores from PC2were interpreted as the relationship between body size(body and leg size) and antenna length.

A factor analysis of variance performed on the scoresfrom PC1 (size of the appendages) showed no effect oflocality but revealed the existence of host-related phe-notypic variation in B. brassicae (Table 2). Aphidsfeeding upon B. campestris showed higher mean scorevalues for PC1 than those of B. oleracea (Fig. 3), indi-cating that the former were larger and with longerappendages. The interaction locality · host was notsignificant (Table 2).

A similar analysis with the scores derived from PC2showed significant effects of locality, host, and theinteraction locality · host (Table 2). The highest averagevalue of the scores from PC2 was observed in aphidsgrowing on B. campestris at Mitziton. In this site and in

Fig. 2 Illustration of the sevenmorphological characters measuredin B. brassicae

Table 1 Principal components analysis of seven morphologicaltraits measured in individuals of B. brassicae in four localities of thehighlands of Chiapas, Mexico

Character PC1 PC2

Body length 0.314 0.770Antennal segment III 0.451 �0.048Hind tarsal segment II 0.378 0.115Base of antennal segment VI 0.339 �0.584Terminal process of lastantennal segment

0.395 �0.110

Hind tibia 0.456 0.061Last rostral segment 0.271 �0.183Eigenvalue 3.489 0.823Percentage of total variance 49.84 11.76

Table 2 F-ratios of mixed analyses of variance of seven morpho-logical characters and of the scores of PC1 and PC2 of B. brassicaepopulations associated to B. oleracea and B. campestris at fourlocalities of Chiapas, Mexico

Character Locality Host Interaction

BL 3.53 NS 6.16 NS 6.17***AIII 16.81* 29.35** 0.40 NSTII 2.02 NS 4.53 NS 5.08**BVI 2.75 NS 1.52 NS 9.18**TP 3.92 NS 0.04 NS 1.97 NSHTIB 2.43 NS 5.49 NS 1.27 NSLRS 204.82*** 46.72*** 0.08 NSPC1 6.41 NS 10.83* 1.95 NSPC2 35.95** 8.52** 14.48***

*P<0.05, **P<0.01, ***P<0.001NS not significant

594

Teopisca, aphids that grew in B. campestris had highervalues of PC2 than those from B. oleracea. The reversewas true at Chamula, thus giving rise to a locality · hostinteraction. No differences between hosts were observedat Balun (Fig. 3).

In order to evaluate the differences among localitiesand hosts in each of the seven morphological charac-ters of B. brassicae, we performed a series of indepen-dent mixed-model analyses of variance with locality asa random factor. Analyses revealed significant differ-ences between hosts and among localities in the thirdantennal segment and in the last rostral segment. Theinteraction locality · host was significant for bodylength, the second tarsal segment of the hind leg, and

for the base of the sixth antennal segment (Table 2,Fig. 4).

Discussion

The results of this study demonstrate that most of themorphological variation displayed by B. brassicae isaccounted for by the host plant used in each localityand, to a lesser extent, by differences among localities.Morphological differentiation between aphids inhabitingdifferent hosts within localities suggests a scenario ofincipient host-driven differentiation. Such a scenario isconsistent with previous studies showing small but sig-nificant genetic differentiation between host-associatedpopulations of B. brassicae (Ruiz-Montoya et al. 2003).Nonetheless, this interpretation needs further supportfrom evidence showing that morphological differencesbetween hosts are genetically based and affected byselection.

Principal components analysis showed that most(50%) of the morphological variation of B. brassicae isrelated to the general size of the appendages. Append-ages of aphids growing on B. campestris were larger inthree out of four localities, suggesting that exploitationof this host species requires larger appendages. Giventhat B. campestris is more pubescent than B. oleraceae,the difference in appendage size may be related to hostusage. In fact, there is some evidence showing that thesize of the appendages is associated with host manipu-lation. For instance, fine, long, hind tarsal segments inaphids are frequently associated with pubescent surfacesof host plants (Moran 1986; Heie 1987). Although ourresults support this interpretation, we do not know if thedifferences in the size of the appendages are driven bythe different selective regimes imposed by host species orif they are simply an expression of phenotypic plasticity.Further investigation should be focused on the fitnessconsequences of reciprocal transfer experiments, whereaphids of a given morphology are exposed to differenthosts.

The relationship between body and antenna size(scores from PC2) showed a marked variation associatedwith the differences among localities and hosts, and theirinteraction. This result suggests that allometric rela-tionships in B. brassicae are influenced by the interactionbetween these two sources of environmental variation soas to produce a complex pattern of aphid morphology.

Results from this study suggest that the two host-plant species exploited by B. brassicae in the highlandsof Chiapas constitute contrasting feeding environmentsthat favour different morphologies. B. campestris andB. oleraceae differ in their nutritional quality anddefensive characteristics. For aphids, nutritional qualityof a host plant is related to the availability of nitrogen,amino acids and sucrose (Sandstrom 1994; Chen et al.1997; Ashford et al. 2000). Aphids reared on nutri-tionally poor hosts (e.g., low content of nitrogen oramino acids) are smaller than those reared on nutri-

Fig. 3 Average values (±1 SE) of the individual scores of theprincipal component 1 (PC1) and 2 (PC2) of B. brassicae inrelation to the host plant used in four localities of Chiapas(Mexico)

595

tionally rich hosts (Dixon and Kindlmann 1994).B. campestris has a higher content of nitrogen thanB. oleraceae (L.Ruiz-Montoya, unpublished data), andthe reproductive success of B. brassicae is superior inB. campestris (Ruiz-Montoya and Nunez-Farfan 2004).Thus, although B. campestris seems to be a better host

for B. brassicae, aphids with long appendages wouldbetter exploit this host. Under this interpretation, B.oleraceae could be considered as a relatively lowquality resource used by aphids whose access to thehigh quality B. campestris is restricted by their shortappendages.

Fig. 4 Average values (±1 SE)of seven morphologicalcharacters of B. brassicae in twohost plants across four localitiesof Chiapas (Mexico)

596

In summary, the spatial distribution and the nutri-tional/defensive differences between these two hostspecies offer a plausible scenario for phenotypic differ-entiation in B. brassicae. These circumstances establish aheterogeneous environment in terms of host quality andsuggest that colonising individuals choosing high qualityhosts would enjoy a selective advantage. Heterogeneousenvironments may promote the evolution of host-spe-cialisation by host-specific selection, or adaptive pheno-typic plasticity (Meyers and Bull 2002). Previous studieshave detected significant genetic differentiation in enzy-matic loci both among localities (Fst=0.13) and betweenhosts within a locality (0.033 £ Fst £ 0.38) for the samepopulations studied here (Ruiz-Montoya et al. 2003).Genetic differentiation between aphids using differenthosts within a locality is not expected under the hypoth-esis of generalist genotypes, suggesting that the observeddifferences may be the result of host specialisation drivenby natural selection. This scenario would be plausible ifgenetic differentiation were correlated with genes affect-ing morphology, and the amount of plasticity in mor-phology were negligible (Schlichting and Pigliucci 1998).

Finally, it is worth noting that both host species arecommonly found in sympatry (the wild B. campestrisgrows at the edges or near cultivated fields ofB. oleracea),and gene flow between hosts is relatively high (Nm=1.86,enzymatic loci; Ruiz-Montoya et al. 2003), suggestingthat host specialisation via local adaptation could belimited. Further studies regarding the presence of geneticvariation among reaction norms, coupled with selectionanalyses in each environment, are needed before weconclude that the genetic and phenotypic differentiationfound in B. brassicae is the product of local adaptation.

Acknowledgements We are grateful to Manuel Giron-Intzın andVianey Mendez-Trujillo for technical assistance. We deeplyappreciate the statistical advice offered by Duncan Golicher. Wethank Martın Aluja, Rodolfo Dirzo, Carlos Cordero, FabianVargas and Shelly Lachys for reading and commenting on themanuscript. CONACyT (31543-B and 990039-DO-LRM), andECOSUR Project 7-Dinamica de Poblaciones y Comunidades deInsectos provided funds for this study. This work is a partial fulf-ilment of the doctoral dissertation of LRM.

References

Agren J, Schemske DW (1994) Evolution of trichome number in anaturalized population of Brassica campestris. AmNat 143:1–13

Anstead JA, Burd JD, Shufran KA (2002) Mitochondrial DNAsequence divergence among Schizaphis graminum (Hemiptera:Aphididae) clones from cultivated and non-cultivated hosts:haplotype and host associations. Bull Entomol Res 92:17–24

Ashford DA, Smith WA, Douglas AE (2000) Living on a highsugar diet: the fate of sucrose ingested by phloem-feeding insect,the pea aphid Acyrthosiphon pisum. J Insect Physiol 46:335–341

Berlocher SH, Feder JL (2002) Sympatric speciation in phytopha-gous insects: moving beyond controversy? Annu Rev Entomol47:773–815

Blackman RL (1987) Morphological discrimination of a tobacco-feeding form from Myzus persicae (Sulzer) (Hemiptera: Aph-ididae), a key to new world Myzus (Nectarosiphon) species. BullEntomol Res 77:713–730

Blackman RL, Eastop VF (2000) Aphids on the world’s crops. Anidentification and information guide, 2nd edn. Wiley, London

Blackman RL, Paterson AJC (1986) Separation of Myzus (Nec-tarosiphon) antirrhinii (Macchiati) from Myzus (N.) persicae(Sulzer) and related species in Europe (Homoptera: Aphididae).Syst Entomol 11:267–276

Bush GL (1994) Sympatric speciation in animals: new wine in oldbottles. Trends Ecol Evol 9:285–288

Carroll SP, Boyd C (1992) Host race radiation in the soapberrybug: natural history with the history. Evolution 46:1052–1069

Chen JQ, van Rhabe Y, Delobel B, Sauvion N, Guillaud J, FebvayG (1997) Melon resistance to the aphid Aphis gossypii:behavioural analysis and chemical correlations with nitroge-nous compounds. Entomol Exp Appl 85:33–44

Damsteegt VD, Voegtlin DJ (1990) Morphological and biologicalvariation among populations of Aulacorthum solani (Homop-tera: Aphididae). Ann Entomol Soc Am 83:949–955

Dixon AFG (1998) Aphid ecology. Chapman & Hall, LondonDixon AFG, Kindlmann P (1994) Optimum body size in aphids.

Ecol Entomol 19:121–126Eastop VF (1985) The acquisition and processing of taxonomic

data. In: Szelegiewicz H (ed) Proceedings of the internationalaphidological symposium, Jablonna, Ossolineum, Wroclaw,Poland, pp 245–270

Fox LR, Morrow PA (1981) Specialization: species property orlocal phenomenon? Science 211:887–893

Gonzalez-Espinosa M, Ochoa-Gaona S, Ramırez-Marcial N,Quintana-Ascencio PF (1997) Contexto vegetacional y florısticode la agricultura en Los Altos de Chiapas. In: Parra-VR, Dıaz-Hernandez B (eds) Agricultura y crisis rural, Tomo I. RecursosNaturales, El Colegio de la Frontera Sur. San Cristobal de lasCasas, Chiapas, Mexico, pp 110–135

Heie OE (1987) Morphological structures and adaptations. In:Minks AK, Harrewijn P (eds) World crop pests, aphids theirbiology, natural enemies and control. Elsevier, Amsterdam, pp393–400

Lazzari SMN, Voegtlin DJ (1993) Morphological variation inRhopalosiphum padi and R. insertum (Homoptera: Aphididae)related to host plant and temperature. Ann Entomol Soc Am86:26–36

Louda S, Mole S (1991) Glucosinolates: chemistry and ecology. In:Rosenthal GA, Berenbaum MR (eds) Herbivores, their inter-actions with secondary plant metabolites, 2nd edn. AcademicPress, Oxford, pp 123–164

Meyers LA, Bull JJ (2002) Fighting change with change: adaptivevariation in an uncertain world. Trends Ecol Evol 17:551–557

Montoya-Gomez G (1998) Participacion de los productorescampesinos en el mercado de hortalizas en Los Altos de Chia-pas: El caso de la papa. In: Reyes-Ramos ME, Moguel-ViverosR, Van der Harr G (eds) Espacios disputados: transformacionesrurales en Chiapas, Universidad Autonoma Metropolitana y ElColegio de la Frontera Sur, Mexico DF, pp 187–209

Moran N (1981) Intra specific variability in herbivore performanceand host quality: a field study of Uroleucon caligatum (Ho-moptera: Aphididae) and its Solidago hosts (Asteraceae). EcolEntomol 6:301–306

Moran N (1986) Morphological adaptation to host plants inUroleucon (Homoptera: Aphididae). Evolution 40:1044–1050

Moran N (1992) The evolutionary maintenance of alternativephenotypes. Am Nat 139:971–989

Nagel L, Schluter D (1998) Body size, natural selection, and spe-ciation in sticklebacks. Evolution 52:209–218

Nosil P, Crespi BJ (2004) Does gene flow constrain adaptivedivergence or vice versa? A test using ecomorphology andsexual isolation in Timema cristinae walking sticks. Evolution58:102–112

Pappers MS, van Dommelen H, van der Velde G, Ouborg NJ(2001) Differences in morphology and reproductive traits ofGalerucella nymphaeae from four plant species. Entomol ExpAppl 99:183–191

Roff DA (1992) The evolution of life histories, theory and analysis.Chapman & Hall, New York

597

Rollins RC (1993) The Cruciferae of continental North America.Stanford University Press, Palo Alto

Rollins RC, Banerjee UC (1976) Trichomes in studies of the Cru-ciferae. In: Vaughan JG, Macleod AJ, Jones BMG (eds) Thebiology and chemistry of the Cruciferae. Academic Press, NewYork, pp 145–167

Ruiz-Montoya L, Nunez-Farfan J (2004) Local adaptation ofBrevicoryne brassicae (Homoptera: Aphididae) to host plant atLos Altos de Chiapas, Mexico. In: Simon JC, Dedryver CA,Rispe C, Hulle M (eds) Aphids in a new millenium, InstitutteNational de la Recherche Agronomique (INRA) Paris, France,pp 261–266

Ruiz-Montoya L, Zuniga-Bermudez G, Pena-Martınez R (1994)Morphological variation of Eriosoma lanigerum (Homoptera:Aphididae) on Pyracantha koidzumii in Mexico City. AnnEntomol Soc Am 87:108–115

Ruiz-Montoya L, Nunez-Farfan J, Vargas J (2003) Host-associatedgenetic structure of Mexican populations of the cabbage aphidBrevicoryne brassicae (Homoptera: Aphididae). Heredity91:415–421

Sandstrom J (1994) Performance of pea aphid (Acyrtosiphon pisum)clones on host plants and synthetic diets mimicking the sameplants phloem amino acid composition. J Insect Physiol40:1051–1057

SAS Institute (1995) JMP statistics and graphics guide. Version 3.1.SAS, Cary

Schlichting CD, Pigliucci M (1998) Phenotypic evolution, a reac-tion norm perspective. Sinauer Associates, Sunderland, MA

Schluter D (1995) Adaptive radiation in sticklebacks: trade-offs infeeding performance and growth. Ecology 76:82–90

Schluter D (2001) The ecology of adaptive radiations. OxfordUniversity Press, London

Shaposhnikov GCH (1985) The main features of the evolution ofaphids. In: Szelegiewicz H (ed) Proceedings of the internationalaphidological symposium, Jablonna, Ossolineum, Wroclaw,Poland, pp 19–99

Sokal RR, Unnasch RS, Thomson BA (1991) Pemphigus revisited:changes in geographic variation but constancy in variability andcovariation. Evolution 45:1585–1605

Stowe KA (1998) Experimental evolution of resistance in Brassicacampestris: correlated response of tolerance in lines selected forglucosinolate content. Evolution 52:703–712

Sultan SE, Spencer HG (2002) Metapopulation structure favorsplasticity over local adaptation. Am Nat 160:271–283

Takada H (1979) Characteristics of forms of Myzus persicae (Sul-zer) Homoptera: Aphididae) distinguished by colour andesterase differences and their occurrence in populations ondifferent host plants in Japan. Appl Entomol Zool 14:349–360

Underwood AJ (1997) Experiments in ecology. Their logical designand interpretation using analysis of variance. Cambridge Uni-versity Press, Cambridge

Via S (1987) Genetic constraints on the evolution of phenotypicplasticity. In: Loeschcke V (ed) Genetic constraints on adaptiveevolution. Springer-Verlag, Berlin Heidelberg, pp 46–71

Via S (2001) Sympatric speciation in animals: the ugly duckinggrows up. Trends Ecol Evol 16:381–390

Via S (2002) The ecological genetics of speciation. Am Nat 159:S1–S7

Via S, Shaw AK (1996) Short-term evolution in the size and shapeof pea aphids. Evolution 50:163–173

Weber G (1985) Genetic variability in host plant adaptation of thegreen peach aphid Myzus persicae. Entomol Exp Appl 38:49–56

Weisser WW, Stadler B (1994) Phenotypic plasticity and fitness inaphids. Eur J Entomol 91:71–78

598