Both host-plant phylogeny and chemistry have shaped the African seed-beetle radiation

For the last 40 years, many authors have attempted to characterize the main patterns of plant-insect evolutionary interactions and understand their causes. In the present work on African seed-beetles (Coleoptera: Bruchidae), we have performed a 10-year field work to sample seeds of more than 300 species of potential host-plants (from the family Fabaceae), to obtain bruchids by rearing. This seed sampling in the field was followed by the monitoring of adult emergences which gave us the opportunity to identify host-plant use accurately. Then, by using molecular phylogenetics (on a combined data set of four genes), we have investigated the relationships between host-plant preferences and insect phylogeny. Our objectives were to investigate the level of taxonomic conservatism in host-plant fidelity and host-plant chemistry. Our results indicate that phylogenetically related insects are associated with phylogenetically related host-plants but the phylogeny of the latter cannot alone explain the observed patterns. Major host shifts from Papilionoideae to Mimosoideae subfamilies have happened twice independently suggesting that feeding specialization on a given host-plant group is not always a dead end in seed-beetles. If host-plant taxonomy and chemistry in legumes generally provide consistent data, it appears that the nature of the seed secondary compounds may be the major factor driving the diversification of a large clade specializing on the subfamily Mimosoideae in which host-plant taxonomy is not consistent with chemical similarity.     

Evolutionary dynamics of host-plant specialization: a case study of the tribe Nymphalini

Two general patterns that have emerged from the intense studies on insect-host plant associations are a predominance of specialists over generalists and a taxonomic conservatism in host-plant use. In most insect-host plant systems, explanations for these patterns must be based on biases in the processes of host colonizations, host shifts, and specialization, rather than cospeciation. In the present paper, we investigate changes in host range in the nymphalid butterfly tribe Nymphalini, using parsimony optimizations of host-plant data on the butterfly phylogeny. In addition, we performed larval establishment tests to search for larval capacity to feed and survive on plants that have been lost from the female egg-laying repertoire. Optimizations suggested an ancestral association with Urticaceae, and most of the tested species showed a capacity to feed on Urtica dioica regardless of actual host-plant use. In addition, there was a bias among the successful establishments on nonhosts toward plants that are used as hosts by other species in the Nymphalini. An increased likelihood of colonizing ancestral or related plants could also provide an alternative explanation for the observed pattern that some plant families appear to have been colonized independently several times in the tribe. We also show that there is no directionality in host range evolution toward increased specialization, that is, specialization is not a dead end. Instead, changes in host range show a very dynamic pattern.     

Phylogeny and host-plant association in the leaf beetle genus Trirhabda LeConte (Coleoptera: Chrysomelidae)

The leaf beetle genus Trirhabda contains 26 described species from the United States and Canada, feeding on host plants from the families Asteraceae and Hydrophyllaceae. In this study, we present a phylogeny for the genus that was reconstructed from mitochondrial COI and 12S rRNA fragments, nuclear ITS2 rRNA, and morphological characters. Both parsimony and mixed-model Bayesian likelihood analyses were performed. Under both methods, the mitochondrial and nuclear partitions support the same backbone phylogeny, as do the combined data. The utility of the molecular data is contrasted with the low phylogenetic signal among morphological characters. The phylogeny was used to trace the evolution of the host-plant association in Trirhabda. The recovered phylogeny shows that although the host-plant association is phylogenetically conservative, Trirhabda experienced one shift to a distantly related host-plant family, 6 shifts between host-plant tribes, and 6 between genera within tribes. The phylogeny reveals that Trirhabda were plesiomorphically adapted to tolerate complex secondary compounds of its host plants and this adaptation is retained in Trirhabda species, as evidenced by multiple shifts from chemically simpler host plants back to the more complex host plants.     

Parallels in the evolution of the two largest New and Old World seed-beetle genera (Coleoptera, Bruchidae)

This study provides the first phylogenetic analysis of a large sample of the two largest genera of seed-beetles, Acanthoscelides Schilsky and Bruchidius Schilsky, which mostly feed on legumes (Fabaceae). The goal of this study was to investigate evolutionary patterns in relation to biogeography and host-plant associations. We used three mitochondrial molecular markers and parsimony and Bayesian inference methods to reconstruct the phylogeny of 76 species. In addition, we critically reviewed host-plant records in the literature for these two bruchid genera. Our results demonstrated the existence of two major clades, one New World and one largely Old World, which generally correspond to the two genera. Yet, current classification of several species is erroneous, so that both genera as currently defined are paraphyletic. We highlighted a strong trend toward specialization (with high taxonomic conservatism in host-plant use) exhibited by the two studied genera. However, we showed the existence of several host shifts during the evolution of this group of bruchids. Our phylogenetic hypotheses and our evaluation of host-plant associations both suggest that the two genera have undergone parallel evolution, as they have independently colonized similar host plants in their respective areas of distribution. Our estimation of divergence times indicated a more ancient origin for bruchids than that suggested by the fossil records. Interestingly, the suggested timing of diversification is consistent with the hypothesis of a radiation that could have occurred contemporaneously with the diversification of their legume hosts.     

Adaptive radiation of gall-inducing insects within a single host-plant species

Speciation of plant-feeding insects is typically associated with host-plant shifts, with subsequent divergent selection and adaptation to the ecological conditions associated with the new plant. However, a few insect groups have apparently undergone speciation while remaining on the same host-plant species, and such radiations may provide novel insights into the causes of adaptive radiation. We used mitochondrial and nuclear DNA to infer a phylogeny for 14 species of gall-inducing Asphondylia flies (Diptera: Cecidomyiidae) found on Larrea tridentata (creosote bush), which have been considered to be monophyletic based on morphological evidence. Our phylogenetic analyses provide strong support for extensive within-host plant speciation in this group, and it demonstrates that diversification has involved numerous shifts between different plant organs (leaves, buds, flowers, and stems) of the same host-plant species. Within-plant speciation of Asphondylia is thus apparently facilitated by the opportunity to partition the plant ecologically. One clade exhibits temporal isolation among species, which may have facilitated divergence via allochronic shifts. Using a novel method based on Bayesian reconstruction, we show that the rate of change in an ecomorphological trait, ovipositor length, was significantly higher along branches with inferred shifts between host-plant organs than along branches without such shifts. This finding suggests that Larrea gall midges exhibit close morphological adaptation to specific host-plant parts, which may mediate ecological transitions via disruptive selection.     

Phylogenetic analyses of DNA and allozyme data suggest that Gonioctena leaf beetles (Coleoptera; Chrysomelidae) experienced convergent evolution in their history of host-plant family shifts

A phylogenetic analysis of the genus Gonioctena (Coleoptera, Chrysomelidae) based on allozyme data (17 loci) and mitochondrial DNA sequence data (three gene fragments, 1,391 sites) was performed to study the evolutionary history of host-plant shifts among these leaf beetles. This chrysomelid genus is characteristically associated with a high number of different plant families. The diverse molecular data gathered in this study are to a large extent congruent, and the analyses provide a well-supported phylogenetic hypothesis to address questions about the evolution of host-plant shifts in the genus Gonioctena. The most-parsimonious reconstruction of the ancestral host-plant associations, based on the estimated phylogeny, suggests that the Fabaceae was the ancestral host-plant family of the genus. Although most of the host-plant shifts (between different host species) in Gonioctena have occurred within the same plant family or within the same plant genus, at least eight shifts have occurred between hosts belonging to distantly related and chemically dissimilar plant families. In these cases, host shifts may have been simply directed toward plant species available in the environment. Yet, given that two Gonioctena lineages have independently colonized the same three new plant families (Salicaceae, Betulaceae, Rosaceae), including four of the same new genera (Salix, Alnus, Prunus, Sorbus), some constraints are likely to have limited the different possibilities of interfamilial host-plant shifts.     

Statistical analysis of unlabeled point sets: comparing molecules in chemoinformatics

We consider Bayesian methodology for comparing two or more unlabeled point sets. Application of the technique to a set of steroid molecules illustrates its potential utility involving the comparison of molecules in chemoinformatics and bioinformatics. We initially match a pair of molecules, where one molecule is regarded as random and the other fixed. A type of mixture model is proposed for the point set coordinates, and the parameters of the distribution are a labeling matrix (indicating which pairs of points match) and a concentration parameter. An important property of the likelihood is that it is invariant under rotations and translations of the data. Bayesian inference for the parameters is carried out using Markov chain Monte Carlo simulation, and it is demonstrated that the procedure works well on the steroid data. The posterior distribution is difficult to simulate from, due to multiple local modes, and we also use additional data (partial charges on atoms) to help with this task. An approximation is considered for speeding up the simulation algorithm, and the approximating fast algorithm leads to essentially identical inference to that under the exact method for our data. Extensions to multiple molecule alignment are also introduced, and an algorithm is described which also works well on the steroid data set. After all the steroid molecules have been matched, exploratory data analysis is carried out to examine which molecules are similar. Also, further Bayesian inference for the multiple alignment problem is considered.     

Temporal variation in herbivore host-plant preference and performance: constraints on host-plant adaptation

Temporal variation in the acceptability or suitability of plant genotypes to an herbivore has seldom been considered as a possible constraint limiting the adaptation of herbivores to particular plant genotypes, or the occurrence of a positive correlation between host-plant preference and offspring performance. In this study, we used data spanning 12 yr from the same 20 clones of goldenrod ( Solidago altissima ) to examine the temporal variation in oviposition preference and offspring performance of a stem-galling fly, Eurosta solidaginis . We found that the stem galler's preference for, or performance on, the different clones was uncorrelated between years of this study. Furthermore, we found that the relative rankings of clones changed by an average of 31% between successive years. We suggest that these consistently high year-to-year fluctuations in preference and performance by E. solidaginis are likely due to environmental factors (e.g., water and nutrient levels, or abundance of interspecific herbivores) that fluctuate over time and are known to differentially affect the acceptability and suitability of clones to herbivores; i.e., genotype×environment interactions. These results are significant because temporal fluctuations in host-plant preference and performance are likely to favor a more generalized diet by herbivorous insects.     

How foraging tactics determine host-plant use by a polyphagous caterpillar

The use of multiple host-plant species by populations of insect herbivores can result from a variety of possible ecological and behavioral mechanisms. An understanding of the foraging mechanisms determining polyphagy in relation to local ecological conditions is therefore essential to understanding the evolutionary ecology of polyphagy. Here, we evaluate patterns of host-plant use by the polyphagous caterpillar Grammia geneura (Lepidoptera: Arctiidae) in relation to host-plant availability and foraging tactics of individuals. Field surveys of caterpillar feeding and plant abundance carried out across several sites, seasons, and years showed that: (1) G. geneura consistently preferred forbs to grasses and woody plants, (2) forb-feeding was opportunistic, supporting the idea that caterpillars sample locally available host-plants, and (3) there were consistent patterns of host-plant use that were not explained by host-plant availability (electivity). An independent set of 7-h observations of 11 caterpillars showed that electivity for a subset of caterpillar-host associations could be explained by variation in the probability of initiating feeding and the average duration of feeding bouts on different hosts but not by variation in the probability of encountering different hosts, thus providing a behavioral basis for the observed variation in host-plant use. The use of detailed foraging tactics by larvae to explain host-plant use at the population level is a novel contribution of this study.     

Phylogeny of a Macaronesian radiation: host-plant use and possible cryptic speciation in Liparthrum bark beetles

The Macaronesian islands are well known for their unique endemic floras of woody plants. Many of these unusual plant groups provide important novel resources for bark and wood boring beetles which breed in dead or moribund parts of their host plants. The bark beetle genus Liparthrum exploits a wide range of unusual host plants and has its highest proportion of species living on the Macaronesian Islands. We used DNA sequences of the mitochondrial Cytochrome Oxidase I gene and the nuclear Elongation Factor 1 alpha gene, and morphological characters, to estimate the phylogenetic relationships among species endemic to these archipelagos, and to trace the evolution of host-plant use. All parsimony and Bayesian analyses of the combined data, and maximum likelihood analyses of the molecular data, showed that species associated with Euphorbia are monophyletic. We also found genetic and subtle morphological evidence for three cases of cryptic speciation in one polyphyletic species associated with different Euphorbia plants, showing that high levels of host specialisation can occur also in insects breeding in older and very dry, dead plant tissues.     

When insects help to resolve plant phylogeny: evidence for a paraphyletic genus Acacia from the systematics and host-plant range of their seed-predators

In this study we use an indirect method to address the issue of the systematics of the large and economically important genus Acacia (Leguminosae, Mimosoideae, Acacieae). We propose the use of host-preference data in closely related insect species as a potentially useful tool to investigate host systematic issues, especially when other approaches yield inconsistent results. We have examined the evolution of host-plant use of a highly specialized group of seed-feeders who predate Acacia — the seed-beetles (Coleoptera, Chrysomelidae, Bruchinae). First, the evolution of host-plant preferences in a large clade of Bruchidius species was investigated using molecular phylogenetics and character optimization methods. Second, the scope of our study was enlarged by critically reviewing the host-plant records of all bruchine genera associated with Acacia . Both morphological and molecular data were used to define relevant insect clades, for which comparisons of host-plant range were performed. Interestingly, the analyses of host-plant preferences from 163 seed-beetle species recovered similar patterns of host-plant associations in the distinct clades which develop within Acacia seeds. Our results clearly support the hypothesis of Acacia being a paraphyletic genus and provide useful insights with reference to the systematics of the whole subfamily as well. This study should also be of interest to those involved in the numerous biological control programs which either already use or aim to use seed-beetles as auxiliary species to limit the propagation of several invasive legume tree species.     

Host-plant leaps versus host-plant shuffle: a global survey reveals contrasting patterns in an oligophagous insect group (Hemiptera,Psylloidea)

Global analyses of interspecific interactions are rapidly increasing our understanding of patterns and processes at large scales. Understanding how biodiversity assembles and functions on a global scale will increasingly require analyses of complex interactions at different ecological and phylogenetic levels. We present an analysis of host-plant associations in the sap-sucking Psylloidea (～3,800 species) using the most comprehensive assemblage of host data for this group compiled from 66?% of published records. Psyllids are known for high levels of host specificity and host switching between related plants at local scales, but a global survey implicates historical processes that are not entirely consistent with those at local scales. In particular, saltationary host switching events appear to have been a key factor explaining the wide but patchy distribution of psyllid host-plants throughout the angiosperm phylogeny. Alternative explanations involving co-diversification with subsequent extinction seem implausible. At the seed plant family level, we compare associations for psyllids with those of their relatives the aphids, but, despite notable differences in biogeographic distributions, find few plant families (2%) that host only psyllids but not aphids, while a much larger percentage (31%) host aphids but not psyllids, and 43% of plant families distributed throughout the plant phylogeny host neither group.     

Evolutionary dynamics of host-plant use in a genus of leaf-mining moths

Abstract. We used nuclear 28S rDNA sequence data to estimate the phylogeny of 77 leaf-mining Phyllonorycter (Gracillariidae) moth species, including all 55 British species, feeding on 44 different plant genera. There was strong support for both the monophyly of Phyllonorycter and the placement of the genus Cameraria as its sister group. Host-plant use was mapped onto the moth phylogeny and investigated statistically in several ways. First, we show that the estimated level of cospeciation between leaf miners and their host plants is not greater than expected by chance, despite the physical intimacy of the association. Nevertheless, the pattern of host-plant use is far from random, with closely related Phyllonorycter species generally feeding on closely related plants. However, although Phyllonorycter species from a given host plant tend to form distinct clades, there is also statistical support for multiple independent colonizations of some host-plant taxa (e.g. the order Rosales and the genus Corylus ). Despite numerous host shifts, most Phyllonorycter species feed on trees and the few species that attack shrubs or herbs have mostly acquired these habits independently. There is also limited evidence that host shifts to herbs are more likely from shrubs than from trees. Similarly, most species mine the lower surface of leaves but the few upper-surface miners have each evolved the habit independently. Consequently, these shifts to new adaptive zones have not led to substantial radiations.     

Data partitions and complex models in Bayesian analysis: the phylogeny of Gymnophthalmid lizards

Phylogenetic studies incorporating multiple loci, and multiple genomes, are becoming increasingly common. Coincident with this trend in genetic sampling, model-based likelihood techniques including Bayesian phylogenetic methods continue to gain popularity. Few studies, however, have examined model fit and sensitivity to such potentially heterogeneous data partitions within combined data analyses using empirical data. Here we investigate the relative model fit and sensitivity of Bayesian phylogenetic methods when alternative site-specific partitions of among-site rate variation (with and without autocorrelated rates) are considered. Our primary goal in choosing a best-fit model was to employ the simplest model that was a good fit to the data while optimizing topology and/or Bayesian posterior probabilities. Thus, we were not interested in complex models that did not practically affect our interpretation of the topology under study. We applied these alternative models to a four-gene data set including one protein-coding nuclear gene (c-mos), one protein-coding mitochondrial gene (ND4), and two mitochondrial rRNA genes (12S and 16S) for the diverse yet poorly known lizard family Gymnophthalmidae. Our results suggest that the best-fit model partitioned among-site rate variation separately among the c-mos, ND4, and 12S + 16S gene regions. We found this model yielded identical topologies to those from analyses based on the GTR+I+G model, but significantly changed posterior probability estimates of clade support. This partitioned model also produced more precise (less variable) estimates of posterior probabilities across generations of long Bayesian runs, compared to runs employing a GTR+I+G model estimated for the combined data. We use this three-way gamma partitioning in Bayesian analyses to reconstruct a robust phylogenetic hypothesis for the relationships of genera within the lizard family Gymnophthalmidae. We then reevaluate the higher-level taxonomic arrangement of the Gymnophthalmidae. Based on our findings, we discuss the utility of nontraditional parameters for modeling among-site rate variation and the implications and future directions for complex model building and testing.     

The phylogenetics and biochemistry of host-plant specialization in Melitaeine butterflies (Lepidoptera: Nymphalidae)

Butterflies in the tribe Melitaeini (Lepidoptera: Nymphalidae) are known to utilize host plants belonging to 16 families, although most host-plant records are from four families. Of the 16 host-plant families, 12 produce secondary plant metabolites called iridoids. Earlier studies have shown that larvae of several melitaeine species use iridoids as feeding stimulants and sequester these compounds for larval defense. I investigate the evolutionary history of host-plant use in the tribe Melitaeini by testing a recent phylogenetic hypothesis of 65 species representing the four major species groups of the tribe. By simple character optimization of host-plant families and presence/absence of iridoids in the host plants, I find that plant chemistry is a more conservative trait than plant taxonomy. The ancestral host plant(s) of the entire tribe most likely contained iridoids and were likely to be in the plant family Plantaginaceae. A major host shift from plants containing iridoids to plants not containing iridoids has happened three times independently. The results show that the evolution of host-plant use in melitaeines has been (and still is) a dynamic process when considering plant taxonomy, but is relatively stable when considering host-plant chemistry.     

Food preference and performance of the larvae of a specialist herbivore: variation among and within host-plant populations

Specialist herbivores are suggested to be unaffected by or attracted to the defense compounds of their host-plants, and can even prefer higher levels of certain chemicals. Abrostola asclepiadis is a specialist herbivore whose larvae feed on the leaves of Vincetoxicum hirundinaria, which contains toxic alkaloids and is unpalatable to most generalist herbivores. The food choice, leaf consumption and growth of A. asclepiadis larvae were studied to determine whether there is variation among and within host-plant populations in their suitability for this specialist herbivore. There was significant variation in food preference and leaf consumption among host-plant populations, but no differences were found in larval growth and feeding on different host-plant populations. A. asclepiadis larvae preferred host-plant populations with higher alkaloid concentrations, but did not consume more leaf material from plants originating from such populations in a no-choice experiment. There was also some variation in food preference of larvae among host-plant individuals belonging to the same population, suggesting that there was variability in leaf chemistry also within populations. Such variation in larval preference among host-plant genotypes and populations may create potential for coevolutionary dynamics in a spatial mosaic.     

Conflicting mitochondrial and nuclear phylogeographic signals and evolution of host-plant shifts in the boreo-montane leaf beetle Chrysomela lapponica

We conducted a phylogeographic study on the cold-adapted leaf beetle Chrysomela lapponica, that feeds on willow or birch, by sampling several populations throughout most of the geographic distribution of the species, and by sequencing for each individual one mitochondrial and two nuclear DNA fragments. Patterns of DNA sequence variation from the mitochondrial and nuclear loci, as displayed in the median-joining networks, appear to display contradicting historical signal: a deep genealogical divergence is observed with the mitochondrial genome between the Alpine population and all other populations found in the Euro-Siberian distribution of the species, that is completely absent with both nuclear loci. We use coalescence simulations of DNA sequence evolution to test the hypothesis that this apparent conflict is compatible with a neutral model of sequence evolution (i.e., to check whether the stochastic nature of the coalescence process can explain these patterns). Because the simulations show that this is highly unlikely, we consider two alternative hypotheses: (1) introgression of the mitochondrial genome of another species and (2) the effect of natural selection. Although introgression is the most plausible explanation, we fail to identify the source species of the introgressed mitochondrial genome among all known species closely related to C. lapponica. We therefore suggest that the putative introgression event is ancient and the source species is either extinct or currently outside the geographic range of C. lapponica explored in this study. The observed DNA sequence variation also suggests that a host-plant shift from willow to birch has occurred recently and independently in each of the three birch-feeding populations. This emphasizes further the relative ease with which these beetles can escape their ancestral host-plant specialization on willow, but shows at the same time that host-plant shifts are highly constrained, as they only occur between willow and birch.     

Systematics and morphological evolution within the moss family Bryaceae: a comparison between parsimony and Bayesian methods for reconstruction of ancestral character states

The Bryaceae are a large cosmopolitan moss family including genera of significant morphological and taxonomic complexity. Phylogenetic relationships within the Bryaceae were reconstructed based on DNA sequence data from all three genomic compartments. In addition, maximum parsimony and Bayesian inference were employed to reconstruct ancestral character states of 38 morphological plus four habitat characters and eight insertion/deletion events. The recovered phylogenetic patterns are generally in accord with previous phylogenies based on chloroplast DNA sequence data and three major clades are identified. The first clade comprises Bryum bornholmense, B. rubens, B. caespiticium, and Plagiobryum. This corroborates the hypothesis suggested by previous studies that several Bryum species are more closely related to Plagiobryum than to the core Bryum species. The second clade includes Acidodontium, Anomobryum, and Haplodontium, while the third clade contains the core Bryum species plus Imbribryum. Within the latter clade, B. subapiculatum and B. tenuisetum form the sister clade to Imbribryum. Reconstructions of ancestral character states under maximum parsimony and Bayesian inference suggest fourteen morphological synapomorphies for the ingroup and synapomorphies are detected for most clades within the ingroup. Maximum parsimony and Bayesian reconstructions of ancestral character states are mostly congruent although Bayesian inference shows that the posterior probability of ancestral character states may decrease dramatically when node support is taken into account. Bayesian inference also indicates that reconstructions may be ambiguous at internal nodes for highly polymorphic characters.     

Evolutionary shifts between host oak sections and host-plant organs in Andricus gallwasps

Abstract.— Gall-inducing insects have especially intimate interactions with their host plants and generally show great specificity with regard to both the host-plant species and the organ (e.g. flower, leaf) galled. However, the relative roles of shifts between host species and between host-plant organs in the diversification of gall-inducers are uncertain. We employ a novel and general maximum-likelihood approach to show that shifts between host-plant organs occur at a significantly greater rate than shifts between host oak sections in European Andricus gallwasps. This suggests that speciation has more often been associated with gall location shifts than with colonization of new host-plant species, and implies that it may be easier for gall-inducers to colonize new plant organs than new plant species.
Andricus gallwasps have complex life cycles, with obligate alternation of sexual and parthenogenetic generations. Our phylogenetic analyses show that a life cycle with both generations galling white oaks (section Quercus ) is ancestral, with a single shift of the sexual generation onto black oaks (section Cerris ) to generate a clade with a novel host-alternating life cycle. This new life cycle provided the opportunity for further speciation, but may have also increased the risk of extinction of one or both generations by the demographic requirement for co-existence of both host-plant groups. In summary, it appears that Andricus gallwasp radiation may be a two-level process. Speciation events often involve shifts in gall location on the same host species. However, there are only so many ways to gall an oak, and rare shifts to new oak sections may contribute greatly to long-term diversification by opening up whole new adaptive zones.     

Genetics of host-plant preference in the comma butterfly Polygonia c-album (Nymphalidae), and evolutionary implications

In the Lepidoptera, sex-linked genes have been found to be of importance for species differences in, for example, host-plant preference, and have been implicated in ecological speciation. Variation within species is typically not sex-linked. However, in the comma butterfly Polygonia c-album (Nymphalidae) an X-linked gene has been found to play a major role in determining differences in host-plant use between two well separated populations. For this reason, we studied the role of sex-linked genes for host-plant preference within a single Swedish population of this species. Three generations of females with known pedigrees were studied in the laboratory, and they were given a choice between Urtica dioica and Salix caprea in flight cages. We found strong variation among females and significant genetic variance for host-plant preference, but no evidence for major importance of sex linkage of host-plant preference on this local scale. To what extent the observed genetic variation was due to additive genes and/or effects of major genes was not clear from the maximum likelihood analysis. In a follow-up study we sampled females over a larger area. We found strong variation among females, but not among localities, suggesting an open population structure with strong gene flow. From the combined stock, a selection experiment was performed over 2 years and six generations. The selection lines diverged after the first generation of selection and remained separate, but did not diverge further, suggesting a low degree of narrow-sense heritability and that the genetic differences may be effects of major genes. We discuss these results in relation to the possible role of genetics in the radiation of the Lepidoptera and other phytophagous insects.  © 2005 The Linnean Society of London, Biological Journal of the Linnean Society , 2005, 84 , 755–765.