New Zealand is currently the only major fruit producing country in the world that is free of economically important fruit flies. As part of the effort to maintain this status, there is a need to supplement quarantine decision-making procedures with a means of rapidly identifying immature life stage infestations to the species level. Here we describe a molecular method that achieves this, using simple restriction patterns of ribosomal DNA (rDNA) as diagnostic markers. The 18S and 18S plus internal transcribed spacer (ITS) regions were amplified from larval DNA by the polymerase chain reaction (PCR). Nineteen species, spanning four genera (including five subgenera of Bactrocera) were analysed. Restriction analysis of the 18S PCR product provided poor resolution, even at the generic level. Digestion of the 18S + ITS PCR product, however, generated thirteen diagnostic haplotypes as defined by the composite restriction patterns from RsaI, Sau3a HaeIII and AluI. No variation was detected at these restriction sites within or between populations. Twenty two restriction enzymes have been screened, but diagnostic RFLPs have yet to been found for six out of the ten Bactrocera (Bactrocera) species; B. passiflorae (Froggatt) cannot be distinguished from B. facialis (Coquillet), nor B. kirki (Froggatt) from B. trilineola (Froggatt) or B. neohumeralis (Hardy) from B. tryoni (Froggatt). Geographic origin could assist in distinguishing the first four species, but the latter pair are very closely related with overlapping origins, hosts and adult morphology. All six species, however, are considered high risk with respect to their likely establishment in New Zealand. Therefore diagnosis based on this molecular technique would support the same quarantine decision. We consider this method could be useful as a diagnostic technique and discuss directions for further development.

Reproduction of Varroa jacobsoni Oudemans has been extensively studied in colonies of the mite's recently acquired host, Apis mellifera Linnaeus. However, data on reproduction in colonies of its original host, A. cerana Fabricius, are still fragmentary. In order to compare reproductive success in naturally infested brood cells of A. cerana and A. mellifera, freshly capped worker and drone cells (0–5 h after capping) were marked and opened 9 days later. When one or more foundress Varroa mites were found, the number and stage of the offspring were recorded. In colonies of A. cerana, V. jacobsoni reproduced almost exclusively in drone cells. Surprisingly, however, these drone cells proved to be better suited for reproduction than drone (and worker) cells of A. mellifera. Firstly, in drone cells of A. cerana, 97–100% of the mites produced daughters and 98–99% of these mites also produced a son. In drone and worker cells of A. mellifera only 70–80% produced daughters and only 74–86% of these mites produced a son. Secondly, if a mite produced daughters, she produced more daughters in drone cells of A. cerana than in drone cells of A. mellifera: 4.5–4.6 vs 2.9–3.7 daughters per mite, respectively. Finally, infestation by two foundress females negatively affected reproduction per mite in drone cells of A. mellifera, but not in drone cells of A. cerana. Thus, V. jacobsoni seems to be better adapted to A. cerana than to A. mellifera, although population growth of mites in A. mellifera colonies is higher due to the exploitation of worker cells besides drone cells.

Three species of Mesoclanis flies that attack the developing seeds of Chrysanthemoides monilifera in South Africa are being assessed for their potential as biological control agents of C. monilifera in Australia. Surveys in South Africa indicated that M. polana Munro is restricted to C. m. rotundata along the subtropical northern coast, M. magnipalpis Bezzi and M. dubia Walker occur on all three subspecies sampled (C. m. rotundata, C. m. monilifera, C. m. pisifera), although M. dubia is less abundant and has a more restricted distribution than M. magnipalpis. Mesoclanis eggs are laid into a wide range of flowerhead stages of C. m. rotundata over a period of approximately 12–14 days, from very small flowerheads with involucral bracts still completely closed, to flowerheads with the laminae of ray florets (petals) starting to elongate. Levels of oviposition by M. magnipalpis into C. m. pisifera are low in very young flowerheads, possibly as a consequence of a heavy layer of indumentum on young C. m. pisifera flowerheads. The pattern and location of feeding and pupation within flowerheads differs slightly for the three Mesoclanis species, but none of the species is associated with the formation of gall tissue. The effect of Mesoclanis activity on ovule development was assessed on four occasions. Mesoclanis destroyed all the ovules in some capitula, and the maximum level of ovule destruction recorded for a site was 64%.

Although Aphytis lingnanensis Compère is an important parasitoid of California red scale Aonidiella aurantii Maskell, several ‘races’ of A. lingnanensis exist which parasitize white louse scale Unaspis citri Comstock. The reproductive status of a white louse scale ‘race’ of A. lingnanensis that originated in Thailand was examined in relation to several populations (from Queensland, California and the Philippines) that parasitize California red scale. None of the wasps from red scale mated with individuals from white louse scale in small cages within 10 min, whereas control crosses all mated in that time. The two ‘races’ would therefore constitute independent reproductive entities (species) in sympatry in the field, although no consistent anatomical differences could be found between them, even with the aid of discriminant function analysis on the number of setae on the delta region of the forewing and the mesoscutum.

Aphytis lingnanensis from California red scale in Queensland mated readily with wasps derived from the same host species in California and the Philippines, and mating took place at random among individuals in mate choice tests. Although such results are equivocal, there are no reasons for suspecting that these different populations of A. lingnanensis from California red scale also comprise more than one species. The results obtained indicate that cross-mating tests designed to quantify reproductive isolation between sexual populations are inappropriate. They should rather be designed to establish whether individuals recognize one another as potential mates. The design of cross-mating tests should therefore consider the usual time to mating, of known conspecifics, under the experimental conditions to be used.

Carbon dioxide, carbon monoxide, chloroform, hydrogen, methane and nitrous oxide emissions were collected from within several mound colonies of the subterranean termite Coptotermes lacteus (Fgottatt) and the surrounding habitat of a dry sclerophyll forest in central Gippsland, Australia, following treatment of the termite mounds with mirex-treated bait blocks. All the mirex-treated colonies died within two weeks of commencing active feeding on the bait blocks. The flux measurements of the gaseous emissions were calculated before, during and after treatment with the mirex-treated decayed wood bait blocks. Carbon dioxide, methane and hydrogen all declined after colony death.

Limburger cheese, previously shown to attract female Anopheles gambiae Giles, was solvent extracted and chemically fractionated into acid and non-acid fractions. The extracts and aliquots of headspace odour of the cheese were analysed by gas chromatography and electron impact mass spectrometry. Nineteen saturated and unsaturated aliphatic fatty acids, ranging in carbon chain length from C2 to C18, were detected. The most abundant acids (>1 mg/g of cheese) identified in the acid extract were ethanoic, propanoic, butanoic, hexadecanoic and 9-octadecenoic acid. The same compounds were identified in analyses of headspace samples but only trace quantities of the less volatile acids (C10 to C16) were present, whilst C18 acids were absent. Behavioural responses of female A. gambiae towards a range of dilutions of the acid extract (in diethyl ether) were recorded in a windtunnel bioassay. The undiluted extract was found to be repellent, but became highly attractive (P «0.001) at lower doses, and was still significantly attractive (P<0.001) when diluted 106times. A synthetic mixture of 12 of the more abundant aliphatic acids identified in the acid extract was found to be significantly attractive (P<0.001) when diluted 108 times. Electroantennographic (EAG) studies showed significant and reproducible responses to (saturated) Limburger cheese headspace. At doses higher than 0.1%, the synthetic mixture of 12 acids elicited significantly higher EAG amplitudes than the solvent control (paraffin oil). EAG responses were recorded from mosquitoes stimulated with C5 to C8 acids, that were characterized by significant dose-dependencies. Weaker, though significant EAG responses were obtained with the less volatile acids (C9 to C14). Only hexadecanoic acid did not elicit a detectable response. The electrophysiological and behavioural responses obtained with fatty acids isolated from Limburger cheese suggests that together they could act as a kairomone for female A. gambiae. The implications of this are discussed together with the occurrence and bacterial production of these compounds on human skin.

This study examines the spatial and seasonal patterning of mitochondrial DNA diversity in French populations of the bird cherry-oat aphid, Rhopalosiphum padi (Linnaeus), on both its primary and secondary hosts. Our results confirm the presence of two major mitochondrial lineages that are generally associated with the breeding system variation (cyclic and obligate parthenogenesis) shown by this species. The strength of this relationship varies regionally, being most evident in the south and west. Cyclically parthenogenetic populations show no significant regional or seasonal genetic divergence reflecting high levels of gene flow, possibly promoted by their obligate host-alternation. However, obligately parthenogenetic populations show a north-south cline in the distribution of the dominant haplotypes. This pattern might result from a selective advantage of some obligately parthenogenetic lineages under cold temperature regime. Alternatively, this cline might be established by a gradient in the intensity of nuclear gene flow between cyclically and obligately parthenogenetic populations mediated by androcyclic males. The discrimination between these possible explanations will require extending analysis to include hypervariable nuclear markers.

Adult olive fruit flies Bactrocera (Dacus) oleae Gmelin were exposed for 24 h to 25 × 10 cm polyester target devices, impregnated with 3% a.i. triflumuron (2-chloro-n-[[[trifluromethoxy]phenyl]amino]carbonyl]benzamide), a chitin synthesis inhibitor, formulated in 10% w/v sugar solution and air-dried. Treated insects were tested for egg hatch inhibition and larval mortality. Inhibition of egg hatch and F1 larval mortality depended on the sex exposed and the elapsed time after exposure. Egg hatch was reduced to 37.6% and F1 larval mortality reached 100% when both exposed sexes were mated. Exposure of one sex prior to mating resulted in a lower level of egg inhibition and larval mortality; but exposure of females was more effective at inhibiting egg hatch and larval survival than exposure of males in these crosses. In flies exposed once, the residual activity of triflumuron decreased gradually, fifteen days post-exposure egg hatch and larval mortality reached 62% and 44% respectively. Target devices aged under natural conditions reduced egg hatch to an average of 39% and inhibited completely larval development for a period of four months.

Field studies on the response of Anopheles gambiae s.l. Giles and Anopheles funestus Giles to tents baited with human odour or carbon dioxide were conducted in south-east Tanzania. Two exit traps and a CDC miniature light trap set beside a bed net were used to sample mosquitoes that entered the tent. Human odour, pumped from an underground pit into a bed net attracted a similar number of mosquitoes as a bed net occupied by a human male. Significantly fewer mosquitoes were caught in a tent into which carbon dioxide (300 ml min-1) was pumped than in a human-odour baited tent (9 and 27% for A. gambiae s.l. and A. funestus respectively). A five-fold increase of the carbon dioxide concentration (to 1500 ml min-1) did not increase the catches of A. gambiae s.l. whereas those of A. funestus were increased to 69% of the catches by human odour. Species identifications of A. gambiae s.l. catches showed that A. arabiensis Patton prevailed and that the proportions of A. arabiensis/A. gambiae s.s. did not differ between treatments. It is concluded that in the indoor situation described, human odour other than carbon dioxide is the principal cue to which these malaria vectors are attracted and that the physical presence of a host and carbon dioxide, when used as a kairomone on its own, accounts for only a minor part of the overall attractiveness of man, particularly for A. gambiae s.l.

Diurnal activity patterns and host preference of Glossina fuscipes fuscipes Newstead were studied in forest and linear habitats along Lake Victoria shore, Kenya. The objective was to identify the preferred hosts of G. f. fuscipes, the emanations of which may be attractive to this species. Hourly catches of flies in biconical traps were related to changes in the weather and the prevalence of hosts in the vicinity of traps. Flies were mainly active between 0800 and 1600 h, with males peaking around 1100 h and females around 1300 h. Activity of flies correlated directly with light intensity and temperature, but indirectly with relative humidity. Humans, livestock and the monitor lizard, Varanus niloticus, were the predominant hosts, although a significant positive correlation with fly catches could only be established with the prevalence of lizards. Blood meal identification by microscopic and serological methods showed that 73—98% of G. f. fuscipes fed on monitor lizards irrespective of host prevalence, season or location. The significance and possible epidemiological importance of the relationship between G. f. fuscipes and monitor lizards are discussed.

A comparative analysis was performed on 18 mitochondrial DNA sequences, or haplotypes, of Lutzomyia (Nyssomyia) whitmani (Antunes & Coutinho) isolated by PCR from 28 individual flies originating from 10 Brazilian locations 150–2500 km apart. A phylogenetic analysis using maximum parsimony indicated support for three to four major lineages (Outgroups were haplotypes of Lutzomyia (Nyssomyia) intermedia (Lutz & Neiva)). One L. whitmani lineage was found in the Atlantic Forest zone of the North East, including the species' type locality, and is distinct from a second monophyletic group of haplotypes located in the drier interior of Brazil, stretching from the Tropic of Capricorn to Teresina just outside Amazonia. This provides no support for a previous hypothesis that L. whitmani sensu stricto is a single form widely-distributed south of Amazonia, and characterized by derived anthropophilic and synanthropic behavioural traits. The ranges of both lineages include populations incriminated in the peri-domestic transmission of Leishmania braziliensis sensu stricto. A third, Amazonian, group of haplotypes was less well-defined. It appears to consist of two sub-lineages which, like the two Leishmania species associated with them, are strictly silvatic and are separated by the Amazon floodplain. The zoophily of the Amazonian populations was not shown to be an ancestral trait. The geographical distributions of the Brazilian lineages are concordant with the broad patterns of bioclimate believed to have persisted since the late Tertiary period. These findings are part of ongoing research on the behavioural genetics of L. whitmani sensu lato aimed at understanding the evolution and maintenance of peri-domestic transmission of human cutaneous leishmaniasis.

Two organophosphate (OP) resistant strains of the cattle tick Boophilus microplus (Canestrini) from Mexico and Costa Rica were used to analyse the presence of esterase activity associated with resistance. The concentrations of six major proteins in both resistant strains were increased compared to the susceptible Morelos strain, both when stained with Coomassie Brilliant Blue after SDS-PAGE, and when analysed for esterase activity by the hydrolysis of naphthyl acetate esters. Esterases were named A or B in relation to the substrate preference for alpha or beta naphthyl acetate and numbered according to their position on the SDS—PAGE. The molecular weights of these proteins were: 125, 115, 108, 77, 43 and 67 Kd for Est-Bl, Est-B2, Est-B3, Est-B4, Est-B5 and Est-A respectively. Est-B3 showed cholinesterase (ChE) activity. This study strengthens the hypothesis that the mechanism associated with OP resistance found in many other insects includes an increase of esterase activity, probably as a result of gene amplification. The genes encoding these enzymes could be potentially used as molecular markers to detect resistance in the cattle tick B. microplus using a DNA probe.

In this paper we describe a semi-automated digital image analysis system which is capable of discriminating five closely related species of Ichneumonidae. Specimens were distinguished by differences in their wings. The system functions by (a) extracting the significant variation (principal components) among a training set of images of the same species, (b) using these principal components to efficiently represent the morphology of wings of that species, and (c) exploiting the fact that images of the same species will share characteristic principal components, while images of different species will not. Such an approach allows the construction of modular species classifiers, to which like species correlate strongly, while dissimilar species do not. A recognition accuracy of 94% was achieved when the system was tested on 175 images of wings of the five ichneumonids. The wing images were caricatured to accentuate their venation and pigmentation patterns.