Cytotaxonomic analysis of the polytene chromosomes from larvae of the Simulium damnosum Theobald complex from the island of Bioko in Equatorial Guinea is reported, and a new endemic cytoform is described. Chromosomally this cytoform is close to both S. squamosum (Enderlein) and S. yahense Vajime & Dunbar, but is not identical to either. However, it is morphologically and enzymatically identical to S. yahense. The Bioko form was also found to differ from other cytoforms of the S. damnosum complex in West Africa in the copy number or RFLP pattern of several different repetitive DNA sequences. It is clear that the Bioko form is genetically distinct from other populations of the S. damnosum complex, and whilst it is closest to S. yahense, it shows features that suggest a high degree of geographical and genetic isolation. Such isolation is an important consideration in the assessment of the potential for onchocerciasis vector eradication on Bioko.

Ninety-nine and 476 clones of Myzus persicae (Sulzer) were sampled on tobacco at Kyoto (Shimogamo) (1996–2000) and at 23 other localities (1998–1999) in Japan, respectively. The clones were classified into colour-esterase forms, distinguished by combinations of body colour and electrophoretically detectable esterases, to verify the changes in genotypic composition during the past two decades. Fifteen and 31 colour-esterase forms were found at Kyoto (Shimogamo) and at 23 other localities, respectively. Fourteen (representing c. 95% of the total clones sampled) and 24 (c. 44%) colour-esterase forms, respectively, were different from those found on tobacco during 1978–1985. The frequency of the green form and the very highly insecticide-resistant form increased. More than half of the colour-esterase forms found in the present survey were newly detected. The factors associated with these changes are discussed.

The cytochrome oxidase subunit I gene in mitochondrial DNA of 53 larvae of Contarinia maculipennis Felt from flower buds of various host plants, collected from Hawaii, Japan and Thailand was analysed. Monophyly of the clade including C. maculipennis from Hawaii, Thailand and Japan was supported. There was no sequential variation within the specimens from Hawaii and Japan, which differed from one another by 6 bp (1.37%). Three haplotypes were recognized in specimens from Thailand but differences from Hawaiian and Japanese specimens were small. Overall, there were no differences in the 146 deduced amino acid residues. It is therefore concluded that C. maculipennis is a polyphagous species that can develop on plant hosts representing at least seven botanical families. This pest of Dendrobium flower buds in glasshouses is considered to have entered Hawaii, Florida and Japan from Southeast Asia, and was recently intercepted in the Netherlands. Infestations have established and spread in orchid glasshouses, causing concern about the possibility of more extensive damage to orchids and to crops, such as bitter gourd, grown in close proximity to orchid glasshouses in Japan. The potential usefulness of DNA analysis in determining host plant ranges of morphologically identical cecidomyiid species that are currently identified solely on differences of host plant is emphasized.

The population density of mealybug species in some South African citrus orchards has increased to pest status in recent years. The characterization of the natural enemy complex and quantification of their contribution to the control of Planococcus citri (Risso), Pseudococcus longispinus (Targioni-Tozzetti) and Pseudococcus calceolariae (Maskell) on Citrus limon (L.) and Citrus reticulata (Blanco) was investigated through intensive sampling. Eight primary and four secondary parasitoids, and two predator species were identified from P. citri and P. calceolariae. Anagyrus pseudococci (Girault) and Coccidoxenoides peregrinus (Timberlake) were the most common species, accounting for 44% and 21% of the total. Of the five primary parasitoids reared from P. longispinus, A. pseudococci and Anagyrus sp. were predominant, comprising 41% and 30%. Nymphal and adult parasitism (range = 0–26% vs. 0–66%) and predation (range = 0–5.6% vs. 0–4.1%) varied significantly between host trees and mealybug species (P < 0.001). The numbers of nymphal instars and adult stages of P. calceolariae and P. longispinus and the nymphal stage of P. citri that were parasitized and killed by predators correlated significantly with the total number of hosts on which they acted (P < 0.01), suggesting a density-dependent association. Laboratory bioassay of nine contact insecticides (methidathion, methomyl, methyl-parathion, parathion, profenofos and prothiofos) against C. peregrinus indicated that all were highly toxic, causing 98–100% mortality in < 6 h of treatment. The IGRs fenoxycarb and triflumuron did not cause significant parasitoid mortality (P > 0.05). However, a mixture of pyriproxyfen and mineral oil caused a marginally significant mortality (P < 0.05).

Random amplified polymorphic DNA (RAPD) analysis was applied to 120 individuals of Marchalina hellenica (Gennadius) representing six populations collected in northern, central and southern mainland Greece. One population was sampled on one species of fir tree and the others on two species of pine trees. Four random decamer primers were used to evaluate genetic variation among the populations examined. The results revealed intra- and interpopulation polymorphism both related to host type and region of origin. Phylogenetic analysis based on genetic distances estimated by the RAPD frequencies revealed an important genetic differentiation in samples collected on fir trees in southern Greece and to a lesser extent in samples from pine trees in central and northern Greece. Furthermore, considerable subdivision and restricted gene flow among the populations examined were observed. The results are discussed in relation to the biology and geographical distribution of M. hellenica in Greece.

Laboratory experiments were conducted to examine host selection by Cotesia plutellae Kurdjumov when larvae of its host, Plutella xylostella (Linnaeus), fed on Chinese cabbage, Brassica campestris L. ssp. pekinensis and those fed on common cabbage, Brassica oleracea L. var. capitata were provided simultaneously, and to investigate the roles of plant and host volatiles in mediating host selection. When C. plutellae were provided with equal numbers of host larvae on plants of the two species in one arena, the parasitoid parasitized 4- to 15-fold more host larvae on Chinese cabbage than on common cabbage. This preference changed little with host density. However, an experience of searching coupled with an oviposition in a host larva on a leaf of the less-preferred plant, common cabbage, significantly increased the preference for parasitizing host larvae on this plant and resulted in twice as many host larvae parasitized on this plant than on Chinese cabbage. Dual choice tests with a Y-tube olfactometer showed that plant volatiles from Chinese cabbage were more attractive to female C. plutellae than those from common cabbage when plants of both species were either intact or infested. In parallel to the increased parasitism on common cabbage following experience, oviposition in a host larva on this less-preferred plant significantly increased the response to volatiles emanating from that plant. These results indicate that host plants may strongly influence the foraging behaviour of C. plutellae, but their differential attractiveness to the parasitoid may be altered by experience of the parasitoid.

The silvicultural management of Scottish birch woodlands for timber production is replacing traditional low intensity management practices, such as domesticated livestock grazing. These new management practices involve thinning of existing woodlands to prescribed densities to maximize biomass and timber quality. Although presently infrequent, the wide scale adoption of this practice could affect invertebrate community diversity. The impact of these changes in management on Staphylinidae and Carabidae (Coleoptera) in 19 woodlands in Aberdeenshire, north-east Scotland was investigated. Grazing and logging practices were important determinants of beetle community structure. Woodland area had no effect on any measure of beetle community structure, although isolation did influence the abundance of one carabid species. Changes towards timber production forestry will influence the structure of invertebrate communities, although the scale at which this occurs will determine its effect.

Peach shoot volatiles were attractive to mated female oriental fruit moth, Cydia molesta (Busck), in a dual choice arena. No preference was observed between leaf odours from the principle host plant, peach, and the secondary host plant, apple. Twenty-two compounds were identified in headspace volatiles of peach shoots using gas chromatography–mass spectrometry. Green leaf volatiles accounted for more than 50% of the total emitted volatiles. A bioassay-assisted fractionation using different sorbent polymers indicated an attractant effect of compounds with a chain length of 6–8 carbon atoms. The major compounds of this fraction were tested either singly or in combinations for behavioural response of females. Significant bioactivity was found for a three-component mixture of (Z)-3-hexen-1-yl acetate, (Z)-3-hexen-1-ol and benzaldehyde in a 4:1:1 ratio. This synthetic mixture elicited a similar attractant effect as the full natural blend from peach shoots as well as the bioactive fraction.

Insect seed predators of 24 dipterocarp species (including the genera of Dipterocarpus, Dryobalanops and Shorea) and five species belonging to the Moraceae, Myrtaceae, Celastraceae and Sapotaceae were investigated. In a tropical lowland dipterocarp forest in Sarawak, Malaysia, these trees produced seeds irregularly but intensely during general flowering and seeding events in 1996 and/or 1998. Dipterocarp seeds were preyed on by 51 insect species (11 families), which were roughly classified into three taxonomic groups: smaller moths (Tortricidae, Pyralidae, Crambidae, Immidae, Sesiidae and Cosmopterigidae), scolytids (Scolytidae) and weevils (Curculionidae, Apionidae, Anthribidae, and Attelabidae). Although the host-specificity of invertebrate seed predators has been assumed to be high in tropical forests, it was found that the diet ranges of some insect predators were relatively wide and overlapped one another. Most seed predators that were collected in both study years changed their diets between general flowering and seeding events. The results of cluster analyses, based on the number of adults of each predator species that emerged from 100 seeds of each tree species, suggested that the dominant species was not consistent, alternating between the two years.

The soybean pod gall midge is an important pest of soybean in Japan and is known to occur also in Indonesia and China. This gall midge is described from Japan as Asphondylia yushimaisp. n. and is clearly distinguished from its congeners by the arrangement of the lower frontal horns of the pupa and the sequence of the mtDNA COI region. It is concluded that Prunus zippeliana Miquel is a winter host of the soybean pod gall midge since haplotypes of the soybean pod gall midge coincide with those of the Prunus fruit gall midge that produces fruit galls on P. zippeliana. In addition, phenological and distributional information on the two gall midges and on their host plants supports the identification of the winter host. In Japan, the soybean pod gall midge overwinters as a first instar in the fruit galls on P. zippeliana and emerges as an adult from the galls in May. In summer and autumn, the soybean pod gall midge has two or more generations in the pods of soybean, Glycine max (L.) Merrill or wild fabaceous and caesalpiniaceous plants. Thus host alternation by A. yushimai is confirmed. This is the second finding of host alternation by a species of Asphondylia, the first instance being that of Asphondylia gennadii (Marchal) in Cyprus.

The effect of plant architecture, in terms of leaf hairiness, and prey spatial arrangement, on predation rate of eggs of the spider mite, Tetranychus urticae Koch, by the predatory mite Phytoseiulus persimilis Athias-Henriot was examined on cut stems of chrysanthemums. Three levels of leaf hairiness (trichome density) were obtained using two different chrysanthemum cultivars and two ages within one of the cultivars. The number of prey consumed by P. persimilis was inversely related to trichome density. At low prey densities (less than ten eggs per stem), prey consumption did not differ in a biologically meaningful way between treatments. The effect of prey spatial arrangement on the predation rate of P. persimilis was also examined. Predation rates were higher in prey patches on leaves adjacent to the release point of P. persimilis, but significantly greater numbers of prey were consumed in higher density prey patches compared to low density patches. The predators exhibited non-random searching behaviour, spending more time on leaves closest to the release point. The implications of these findings for biological control and predator–prey dynamics are discussed.

In the UK, Panolis flammea (Denis & Schiffermüller) is a pest of monocultures of non-native lodgepole pine Pinus contorta Douglas, but not of the indigenous host Scots pine P. sylvestrisL. This difference in population dynamics may be due to the adaptation of P. flammea populations to the phenology, chemical composition and natural enemy complement of lodgepole pine. To ascertain if there was local adaptation of P. flammea populations to lodgepole pine, this study tested for improved performance of both larvae and adults on the host plant species from which they were sourced, compared with their performance on the alternative host plant species. No difference was found in the relative mean performance of populations sourced from Scots pine or lodgepole pine plantations, when fed on Scots or lodgepole pine foliage. Larvae grew faster on Scots pine but this difference did not translate into differences in pupal weight, female body weight or fecundity. Indeed, those insects that had fed on lodgepole pine had a longer lifespan than those that had fed on Scots pine, which, if translated into greater probability of mating or higher fecundity, could contribute to the observed outbreak dynamics in the field. The prediction that the observed outbreak dynamics of P. flammea can be explained by the existence of populations locally adapted to lodgepole pine was not supported. These results cast doubt on the use of larval growth parameters as surrogates of fitness in Lepidoptera.

Using a large set of microsatellites, the genetic relationships between three closely related Australian fruit fly species, Bactrocera tryoni (Froggatt), B. neohumeralis (Hardy) and B. aquilonis(May) were investigated. Bactrocera tryoni and B. neohumeralis are sympatric, while B. aquilonis is allopatric to both. The sympatric species, B. tryoni and B. neohumeralis, were found to be genetically distinct. It is likely that despite differences in mating time between these two species, some gene flow still occurs. In contrast, the sibling species B. tryoni and B. aquilonis were found to be closely related, despite allopatry. The level of genetic divergence was similar to that found within eastern Australian populations of B. tryoni. Consideration of all available genetic data suggests that this similarity is not due to recent (i.e. within the last 30 years) displacement of B. aquilonis by B. tryoni from the B. aquilonis region (north-western Australia). Instead the data suggests that, at least in the areas sampled, asymmetrical hybridization may have occurred over a longer timescale.

The genus Diabrotica includes a large number of pest species, including some of the most important crop pests of the Americas. The parasitoid Centistes gasseni Shaw is the first braconid to be described parasitizing Diabrotica in South America, and high natural infestations are reported. Field and experimental observations on the host range, distribution and biology of this parasitoid are described. Centistes gasseni was collected in southern Brazil, eastern Paraguay and northeastern Argentina, in a region comprising humid lowlands and highlands, and cool temperate to warm subtropical climates, with regular rainfall in excess of 1300 mm. Three Diabrotica species, D. limitata (Sahlberg), D. speciosa (Germar) and D. viridula (Fabricius) were found to host the parasitoid, with mean percent parasitism of 5.4, 2.0 and 1.0%, respectively. Diabrotica speciosa and D. viridula are the two most important pest Diabrotica species in South America. Laboratory experiments with field-collected beetles and parasitoid cocoons indicated that C. gasseni overwinters in adult host beetles, remaining dormant in its live host below developmental temperatures. A potential distribution of C. gasseni in North America is proposed based on its known climatic range and the distribution of the main pest species of adult overwintering North American Diabrotica.

During the early stages of the development of termite baits, dyed paper was placed in specified feeding stations to ascertain whether a slow-acting toxicant could be placed in a few bait stations to be delivered to the entire colony members of the Formosan subterranean termite, Coptotermes formosanus Shiraki. Feeding frequency data, as measured by the dye concentration in individual termites, suggested the absence of feeding site fidelity. However, these results were often misinterpreted as random movement of termites that were marked and released for population estimate studies, or the random search of food in soil by subterranean termites. A computer simulation program was constructed to re-examine this feeding frequency data, and confirmed the earlier conclusion that the lack of feeding site fidelity was the most likely explanation for the data.