The effect of the ectoparasitic mite, Varroa destructor on adult worker honeybee (Apis mellifera) emergence weights, water, protein, carbohydrate, and lipid levels

Wet weight, dry weight and water contents of emerging honeybees (Apis mellifera L. [Hymenoptera: Apidae]) infested with the ectoparasitic mite Varroa destructor (Anderson) (Acari: Varroidae) were all negatively correlated with increasing numbers of mites. It was estimated that for every female mite present during the bees' development, the host would lose three percent of its body water. Parasitised bees also emerged with lower head and abdomen concentrations of protein and with lower abdominal carbohydrate concentrations. Lipid concentrations were not detectably affected by V. destructor infestation. The losses of metabolic reserves were not, however, judged to be serious enough to be directly responsible for the high bee mortality and ultimate colony collapse that are associated with the arrival of Varroa in a hive. Some 8.5% of the emerging bees exhibited morphological deformities and deformity was positively correlated with increasing numbers of mites in brood cells. Deformed bees were, however, found in all categories of parasitosis, suggesting that other factors, such as infectious agents, may be involved. Mites that fed on either live or dead U¹⁴C- labelled bees acquired the label within 24 h and it was calculated that an adult female mite consumes 0.67 l haemolymph 24 h^–1. It was also demonstrated that ¹⁴C was transmitted to a previously non-radio-labelled bee when a mite that had been feeding on a labelled bee changed hosts. The level of transfer was above that which could have arisen through contamination of the mites' mouthparts and supports the suggestion that Varroa is an important vector of pathogens such as viruses.     

Effect of acaricide resistance on reproductive ability of the honey bee mite Varroa destructor

The reproduction of pyrethroid-resistant Varroa destructor mite, a brood parasite of honey bees, was observed in Weslaco, Texas, and the results compared with known susceptible mite populations from other studies. Seven Apis mellifera colonies that had mite populations resistant to the acaricide Apistan^™ were used. Pyrethroid-resistance was confirmed when only 17% rather than 90% of mites confined in dishes containing Apistan^™ died after 12 h of exposure. The average number of eggs laid by resistant mites invading worker and drone cells was 4.4 and 5.4 respectively. This is similar to the number of eggs laid by susceptible mites in worker (4.4–4.8) or drone (4.7–5.5) cells. Also the average number of fertilised V. destructor female mites produced by resistant mites in worker (1.0) and drone (2.1) cells were similar to the number produced by susceptible mites in worker (0.9) and drone (1.9–2.2) cells. In addition, no major differences between the resistant and susceptible mite populations were observed in either worker or drone cells when six different reproductive categories and offspring mortality rates were compared. Therefore, it appears that there is little or no reproductive fitness cost associated with pyrethroid resistance in V. destructor in Texas. This revised version was published online in July 2006 with corrections to the Cover Date.     

Feeding Life Style of Redlegged Earth Mite, Halotydeus Destructor (Acari: Penthaleidae), in pastures and the role of broad-leafed weeds

Redlegged earth mite, Halotydeus destructor, is a major pest of pastures and crops in Australia, and also feeds on lower plants on the soil surface. Feeding behaviour is reviewed, and the role of Arctotheca calendula, capeweed, in determining occurrence and abundance in pastures is investigated. Mites fed more and produced more progeny on Trifolium subterraneum (subclover) than on capeweed in non-choice experiments. In pastures with mixed species three times more mites were feeding on subclover than on capeweed foliage. However, twice as many mites were found on the soil surface under pasture patches consisting mainly of capeweed than of subclover. Patches were selected on the basis of pasture height, irrespective of plant composition. Twice as many mites were found in populations under patches of tall than short pasture. The daytime relative humidity was higher in tall than short patches, and the temperature slightly lower, making tall patches a more favourable niche for these mites. Ninety per cent of H. destructor were on the soil surface, while 10% were feeding on the upper canopy of pasture. The proportion of each population that was feeding was greater on subclover than capeweed, and subclover was a more suitable food. In Australia H. destructor occurs in regions with mixed pasture species, grown in rotation with grain crops. The ability of H. destructor to utilize foliage of a range of plant species of differing suitability for food, while living mostly on the soil surface in niches favourable for survival, has enabled it to become very abundant.     

Biocontrol of the shore fly Scatella tenuicosta with Hypoaspis miles and H. aculeifer in peat pots

Hypoaspis miles Berlese and H. aculeiferCanestrini (Acari: Laelapidae) were usedagainst Scatella tenuicosta Collin(Diptera: Ephydridae) at the rates of 7, 14and 36 mites per 0.3-l peat pot (1000, 2000 and5000 per net-m² of peat) with mint (Mentha piperita L.) as the crop. The predatorswere applied either one day (preventivecontrol) or nine days after (curative control)exposing the pots to the flies in thegreenhouse. Based on the number of flies thatemerged from the pots, the control efficacy of preventive H. aculeifer treatments at therate of 36 mites per pot was 84–100% and 92–97% (range over four replicate blocks)after two and three weeks, respectively, fromthe application. The control efficacy ofcurative treatments was 93–100% and 83–94% after two and three weeks, respectively,from the application. Preventive treatmentswith H. miles applied at the rate of 36per pot controlled the flies by 92–100% and 68–97% two and three weeks, respectively, afterthe application. Curative treatments with 36H. miles per pot were less effective (–53–68% and –16–65%, respectively). All treatments with lower mite rates with eitherspecies were ineffective. Due to moreconsistent reduction of fly numbers by Hypoaspis aculeifer, the species was concluded to be a better biocontrol agent ofshore flies in short-term greenhouse cropsgrown in peat pots, probably due to its higherpredation capacity in comparison to H. miles.     

Accumulation and turnover of 2-tridecanone in Tetranychus urticae and its consequences for resistance of wild and cultivated tomatoes

In this study we assessed the dynamic changes of 2-tridecanone in a herbivorous mite (Tetranychus urticae) on tomato (Lycopersicon esculentum, cv. Moneymaker), a plant with methyl ketones in the tetracellular tips of the glandular trichomes (Type VI). We showed that spider mites accumulate 2-tridecanone when foraging on cultivated tomato. Thus, the rate of mite–trichome contact multiplied by the amount of toxin per trichome tip exceeded the relative rate of toxin turnover multiplied by the amount of toxin per mite. The relative rate of toxin turnover was estimated to be 1.1 per day on cucumber, a plant without this toxin. The amount per trichome tip varied from 0.33 ng for middle-leaf trichomes to 1.26 ng for main-stem trichomes. Hence, to achieve a static level of 2-tridecanone equal to 8–17 ng per mite – representing the level we found in mites on middle leaves – the rate of mite–trichome contact should be 26–57 per day. Because methyl ketone apparently accumulates in the spider mites on tomato, the rate of mite–trichome contact is probably higher than that. We expect the accumulation of ketones to occur especially on the stems of cultivated tomato, since this is the area most densely occupied with glandular hairs and because here the hairs have higher levels of the methyl ketones.Using dose–response relationships assessed earlier (Chatzivasileiadis and Sabelis, 1997, 1998), we estimated that the number of mite–trichome contacts causing 50% mortality per day is equal to 88 on a tomato stem, whereas it equals 70 for another strain of spider mites collected from cucumber. On wild tomato, L. hirsutum f. glabratum (PI 134417), just one to two contacts would suffice to cause 50% mortality per day. We suggest that methyl ketones from glandular hairs on tomato are an important mortality factor for spider mites on wild tomato and probably also on cultivated tomato.     

Effectiveness of the BT mite trap for detecting the storage mite pests, Acarus siro, Lepidoglyphus destructor and Tyrophagus longior

Traps have been used extensively to provide early warning of hidden pest infestations. To date, however, there is only one type of trap on the market in the U.K. for storage mites, namely the BT mite trap, or monitor. Laboratory studies have shown that under the test conditions (20 °C, 65% RH) the BT trap is effective at detecting mites for at least 10 days for all three species tested: Lepidoglyphus destructor, Tyrophagus longior and Acarus siro. Further tests showed that all three species reached a trap at a distance of approximately 80 cm in a 24 h period. In experiments using 100 mites of each species, and regardless of either temperature (15 or 20 °C) or relative humidity (65 or 80% RH), the most abundant species in the traps was T. longior, followed by A. siro then L. destructor. Trap catches were highest at 20 °C and 65% RH. Temperature had a greater effect on mite numbers than humidity. Tests using different densities of each mite species showed that the number of L. destructor found in/on the trap was significantly reduced when either of the other two species was dominant. It would appear that there is an interaction between L. destructor and the other two mite species which affects relative numbers found within the trap.The British Crowns right to retain a non-exclusive, royalty-free licence in and to any copyright is acknowledged.This revised version was published online in May 2005 with a corrected cover date.     

Biological control of strawberry tarsonemid mite Phytonemus pallidus and two-spotted spider mite Tetranychus urticae on strawberry in the UK using species of Neoseiulus (Amblyseius) (Acari: Phytoseiidae)

Two species of Neoseiulus, N. californicus and N. cucumeris, showed potential for biocontrol of phytophagous mites on strawberry. N. californicus controlled Tetranychus urticae on potted strawberry plants in a gauze-sided glasshouse at temperatures comparableto early summer in the UK (8–20°C). Both species of phytoseiid reducednumbers of the tarsonemid Phytonemus pallidus on potted strawberry plants under glasshouse conditions (15–23°C). In several experiments reductions in the range of 71–81% in numbers of tarsonemid active stages and eggs, compared to non-release plants, were obtained. The importance of establishing a suitable predator: prey ratio at an earlystage was demonstrated in an experiment where an initial ratio of 1 N. cucumeris: 10 P. pallidus gave a greater degree of controlthan 1:20 or 1:40.     

Ontogenesis of the mite Varroa jacobsoni Oud. in drone brood of the honeybee Apis mellifera L. under natural conditions

A study carried out during the summer of 1994, in southern England, investigated the developmental times and mortality ofVarroa jacobsoni inApis mellifera drone cells. The position and time of capping of 2671 naturally infested drone cells were recorded. Six hours after the cell was capped, 90% of the mites were free from the brood food to start feeding on the developing drone. The developmental time of the mite's first three female offspring (133±3 h) and the male offspring (150 h) and the intervals between egg laying (20–32 h) were similar to those found in worker cells. However, the mortality of the offspring was much lower in drone cells than worker cells. The mode numbers of eggs laid were six and five in drone and worker cells, respectively. All offspring had ample time to develop fully in drone cells with the sixth offspring reaching maturity approximately 1 day before the drone bee emerged. Normal mites (those which lay five or six viable eggs) produced on average four female adult offspring but since only around approximately 55% of the mite population produced viable offspring the mean number of viable adult female offspring per total number of mother mites was 2 to 2.2 in drone cells.     

Varroa destructor reproduction during the winter in Apis mellifera colonies in UK

The reproductive behaviour of female Varroa destructor mites invading worker brood cells during the winter months (January to mid-March) was investigated in four Apis mellifera colonies in UK. The number of viable offspring produced during a reproductive cycle, per mite, was only 0.5 during winter compared with 1.0 during the summer. This was mainly due to a large increase in the population of non-reproductive mites (winter 20%, summer 8%). This increase can be explained by the high level of male offspring mortality observed in winter (42% vs. 18% in summer), which results in nearly half of the newly reared female mites being unfertilised. Since mites that do reproduce lay a similar number of eggs in winter ( ) as in summer ( ), and the level of mortality suffered by the first female offspring is similar in winter (7%) as in summer (6%), it is probably not the internal physiological state of the host which causes the high level of winter non-reproduction, as has been previously suspected.     

Antixenotic resistance of subterranean clover cotyledons to redlegged earth mite, Halotydeus destructor

The short-term response of redlegged earth mite, Halotydeus destructor (Tucker) (Acarina: Penthaleidae) to cotyledons of different varieties of subterranean clover (Trifolium subterraneum subsp. subterraneum L.) was assessed by means of paired choice tests, and no-choice tests. H. destructor had lower numbers and fed less on detached cotyledons of resistant than susceptible varieties, yielding a correlation between the numbers of mites and feeding damage to the cotyledons during a three hour test period. For a number of resistant and susceptible varieties, there was a negative correlation between cotyledon deterrence in the three hour choice test and feeding damage to seedling after a two week period. Since the response of the mites to different subclover varieties occurred within three hours, it is concluded that the resistance is based on antixenosis.No evidence was obtained for antifeedant activity in organic solvent extracts from the variety DGI007 (resistant) in comparison with those from the variety Dalkeith (susceptible). Water soluble compounds from DGI007 cotyledons were preferred by mites, in feeding tests in terms of numbers, over those from Dalkeith (susceptible). Squeezed sap from the cotyledons of both varieties showed the same effects on mites as 5% glucose and were more phagostimulatory than water extracts. Mechanically damaged cotyledons of Dalkeith and DGI007 attracted more mites than the undamaged counterparts. The toughness of cotyledons in 17 varieties of T. subterraneum subsp. subterraneum was measured with a manual penetrometer. Results showed a negative correlation between toughness values and mite feeding damage scores (r²=0.752) for all varieties except S3615D (resistant). This implies a likely involvement of epidermal toughness as a contributor in the antixenotic resistance of these varieties. Other mechanisms may be involved in the resistance of S3615D.     

Life history and reproductive potential of the agarophyte Gelidium robustum in California

The reproductive potential of the tetrasporangial phase of Gelidium robustum was studied for 16 months at two sites off Santa Barbara, California. In all samples tetrasporangial thalli were always more abundant than gametangial ones. Tetratrasporangial sori were present throughout the duration of the study but relative fecundity was highest [300–400 sori g^–1 (w. wt)] in spring/summer samples of consecutive years, as a result of increasing numbers both of tetrasporangial branchlets per plant and of sori per branchlet. On the other hand, laboratory experiments showed that tetraspore release per sorus was highest (150–250 spores sorus^–1 d^–1) in winter. Inferring from these field and laboratory data plants released up to ± 34 000 tetraspores g^–1 (w. wt) d^–1 in the spring/summer of the second study year. Tetraspore germination, under defined culture conditions, also showed a marked seasonality increasing sharply from less than 10% in winter up to almost 60% in spring/summer, thus coinciding with the period of maximal spore output per plant. These results suggest that although relatively high numbers of tetraspores may be released by G. robustum plants all year round these might not always have the potential to germinate and recruit.     

Host plant species and carbohydrate supplements affecting rate of multiplication of redlegged earth mite

The effects of young plants of different legume species on the rate of multiplication of Halotydeus destructor (Tucker)(Acarina: Penthaleidae) were examined in one, two or four week experiments in the laboratory. Rates of multiplication of H. destructor were substantially greater on Vicia sativa (common vetch), Phaseolus vulgaris (dwarf bean) and Pisum sativum (snow pea), than on subterranean clover, Trifolium subterraneum; cultivars Dalkeith, Geraldton and Junee. Levels of soluble carbohydrate and soluble nitrogen were greater in vetch, bean and pea, than in the subclovers. Concentrations of 20 amino acids were not associated with differences in mite performance; pea contained higher levels of glutamine and homoserine than the other plants, which may have adversely affected mite multiplication.In further experiments a carbohydrate supplement caused a fourfold increase in rate of multiplication on subclover cv Junee, but had no effect on rate of multiplication on vetch cv Blanchefleur. The carbohydrate supplement did not cause the rate of multiplication on subclover to equal that on vetch. Subclover is not an optimum host plant for H. destructor nutritionally. The nutritional value of host plants is one factor affecting mite multiplication.     

First detection of <Emphasis Type="Italic">Varroa destructor</Emphasis> resistance to coumaphos in Argentina

In Argentina, studies on Varroa destructor resistance to coumaphos are still unknown. At present, high infestation levels of V. destructor are being detected in colonies of Apis mellifera after treatment with this acaricide. The aim of the present study was to determine the LC₅₀ of coumaphos in V. destructor from four apiaries with high mite density after treatment with coumaphos. The LC₅₀’s were 112, 319, 127 and 133 μg/Petri dish for mites from the four apiaries. Significant LC₅₀ differences were detected between resistant and susceptible mites. LC₅₀ increased 197–559-fold when compared to the corresponding baseline, suggesting the development of resistance. These results are the first report of resistance to coumaphos in V. destructor in Argentina.     

Biological studies including population growth parameters of the predatory miteAmblyseius barkeri [Acarina.: Phytoseiidae] at 25°C in the laboratory

Amblyseius barkeri (Hughes) fed onThrips tabaci (Lind.) at 25°C showed an average duration of 2.2, 0.8 and 3.2 days for the egg, larval and nymphal stages, with mortalities at 1.0, 1.0 and 3.1%, respectively. Females represented 63% of the population and required multiple matings for optimal fertility. The oviposition period was 20.3 days and the average oviposition rate 2.3 eggs per day. The intrinsic rate of increase was 0.22 per day. The expected life span was 29.6 days for ♀♀ and 27.4 days for ♂♂.A. barkeri ♂♂ and ♀♀ both consumed 3.3 nymphs of thrips per day (mean value for the feeding stages). The larva does not take up food. In the absence of thripsA. barkeri was able to consume two-spotted spider mites,Tetranychus urticae (Koch), and their eggs, adult broad mites,Polyphagotarsonemus latus (Banks), and pollen of various plants. Cannibalism was observed when food was lacking. Certain morphological features, egglaying, mating and predatory behaviour are described.      

A laboratory study of feeding and assimilation in Euchlanis dilatata lucksiana

Feeding of Euchlanis dilatata lucksiana, a brachionid rotifer isolated from the Loosdrecht Lakes (The Netherlands), was examined in the laboratory using ¹⁴C-labelled food. The gut-filling time at a food concentration of 9.6 µg C ml^–1 was about 15 minutes. Animals which were fed on 3 size fractions of the lake seston (< 7 µm, 7–15 µm, and 15–33 µm) showed a clear preference (Jacobs' selectivity index) for the largest size fraction. This fraction was composed predominantly of filamentous cyanobacteria. For animals weighing 0.37–0.49 µm C ind.^–1 the daily ration (daily food consumption per unit body weight × 100) ranged from 50 to 100% at food levels of 2 mg C l^–1 and below, but increased to 150–250% at food concentrations of 5 mg C l^–1 and above. The assimilation efficiency was 100% up to 5 mg C l^–1 of food, but decreased to about 80% at higher food levels.     

The ethylene-inhibitor aminoethoxyvinylglycine restores normal nodulation by Rhizobium leguminosarum biovar. viciae on Vicia sativa subsp. nigra by suppressing the ‘Thick and short roots’ phenotype

Nodulation of Vicia sativa subsp. nigra L. by Rhizobium bacteria is coupled to the development of thick and short roots (Tsr). This root phenotype as well as root-hair induction (Hai) and root-hair deformation (Had) are caused by a factor(s) produced by the bacteria in response to plant flavonoids. When very low inoculum concentrations (0.5–5 bacteria·ml^-1) were used, V. sativa plants did not develop the Tsr phenotype and became nodulated earlier than plants with Tsr roots. Furthermore, the nodules of these plants were located on the primary root in contrast to nodules on Tsr roots, which were all located at sites of lateral-root emergence. The average numbers of nodules per plant were not significantly different for these two types of nodulation. Root-growth inhibition and Hai, but not Had, could be mimicked by ethephon, and inhibited by aminoethoxyvinylglycine (AVG). Addition of AVG to co-cultures of Vicia sativa and the standard inoculum concentration of 5·10⁵ bacteria·ml^-1 suppressed the development of the Tsr phenotype and restored nodulation to the pattern that was observed with very low concentrations of bacteria (0.5–5 bacteria·ml^-1). The delay in nodulation on Tsr roots appeared to be caused by the fact that nodule meristems did not develop on the primary root, but only on the emerging laterals. The relationship between Tsr, Hai, Had, and nodulation is discussed.Abbreviations AVG aminoethoxyvinylglycine - cfu colonyforming units - Had root-hair deformation - Hai root-hair induction - NB naringenin-bacteria filtrate - Tsr Thick and short roots     

Trophic determinants of hypopus induction in the stored-product mite Lepidoglyphus destructor (Acari: Astigmata)

A unique (synapomorphic) characteristic of astigmatic mites is the heteromorphic deuteronymph also called hypopus. It is a non-feeding and facultative instar between protonymph and tritonymph. The hypopus is adapted for dispersal and sometimes also for dormancy, as in Lepidoglyphus destructor. The experiments reveal a correlation between the composition of the foodstuff, the duration of development of homomorphic instars, the mortality of protonymphs and the production of hypopodes. As food quality decreases, development lasts longer, mortality increases and hypopodes are produced in greater numbers. Disadvantageous trophic conditions of varied chemical nature favour the induction of hypopodes. The experimental data show that hypopus incidences (as percentage individuals of a population) depend on the relative proportions of constituents of an ingested foodstuff. What matters is the ratio between nourishing foodstuff components and those that are of little or no nutritional value. When a certain ratio does not meet a presumed metabolically required level of nutrients a nutritional deficiency results and hypopus induction is triggered, provided that adequate genetic propensities for hypopus production are present (L. destructor is highly polymorphic for hypopus production). Specific key substances are apparently not involved, and composite properties of a foodstuff are crucial for hypopus induction. Decrease of food quality (not poor food per se) during the hypopus-inducible period (late larval to early protonymphal phase) promotes hypopus induction. The interpretation matches the ecological scene. When trophic deterioration of a patch habitat sets in, often as a result of overcrowding, conditions will eventually become untenable. As a response to incurring nutritional deficiencies the mites will induce hypopodes, which provide for escape from or survival at the decaying habitat patch. Experiments support the threshold model of quantitative genetics for hypopus expression as previously inferred from other experiments with L. destructor.     

The ectoparasitic mite Tropilaelaps mercedesae (Acari, Laelapidae) as a vector of honeybee viruses

The ectoparasitic mites Varroa destructor and Tropilaelaps mercedesae share life history traits and both infect honeybee colonies, Apis mellifera. Since V. destructor is a biological vector of several honeybee viruses, we here test whether T. mercedesae can also be infected and enable virus replication. In Kunming (China), workers and T. mercedesae mites were sampled from three A. mellifera colonies, where workers were exhibiting clinical symptoms of deformed wing virus (DWV). We analysed a pooled bee sample (15 workers) and 29 mites for the presence of Deformed wing virus (DWV), Black queen cell virus (BQCV), Sacbrood virus (SBV), Kashmir bee virus (KBV), Acute bee paralysis virus (ABPV), and Chronic bee paralysis virus (CBPV). Virus positive samples were analysed with a qPCR. Only DWV +RNA was found but with a high titre of up to 10⁸ equivalent virus copies per mite and 10⁶ per bee. Moreover, in all DWV positive mites (N= 12) and in the bee sample virus–RNA was also detected using RT-PCR and tagged RT-PCR, strongly suggesting virus replication. Our data show for the first time that T. mercedesae may be a biological vector of DWV, which would open a novel route of virus spread in A. mellifera. Received 6 June 2008; revised 14 August 2008; accepted 10 September 2008.     

Factors affecting progeny production of Anaphes iole

Anaphes iole Girault(Hymenoptera: Mymaridae) is a native, solitaryegg parasitoid of Lygus spp.(Heteroptera: Miridae) in North America. Current research is considering factors thatmay optimize the in-vivo rearing of A.iole using Lygus hesperus Knight(Heteroptera: Miridae) as host. The effects ofhost density, day of oviposition and foodpresence, parasitoid age and mate presence onproduction of A. iole progeny weredetermined in this study. After exposingindividual parasitoids to host patches for 24h, the percentage of hosts containing a latestage A. iole pupa wassignificantly greater at a moderate hostdensity (41–70 eggs per patch) than at a lowdensity (10–40 eggs). But, no differenceswere detected between moderate and high density(71–100 eggs) or between high and low densitytreatments. More adult progeny were generatedby females (of variable age; 0–2 d old) onthe first day of oviposition rather than thesecond day, regardless of food presence. Progeny sex ratio was decidedly male-biased onthe second day. Female age (0 d vs 1 d old)had a marginal effect on production in 24 h; 0d old females tended to generate more adultprogeny than 1 d old females. Overall, thisresearch suggests that exposing newly-emerged,unfed, mated A. iole females to amoderate to high host density for 1 to 2 dcould lead to time-efficient production ofadult progeny in an in-vivo rearing system.     

Resource limitation of diving ducks at Myvatn: Food limits production

We monitored numbers and production of diving ducks on Lake Myvatn and its outlet river, the Laxá, over a period of 25 years, 1975–1999. Aquatic insects, a major food source, were monitored for 23 years, 1977–1999, using window traps. In this paper, we focus on the relationship between the demography of five species of diving ducks and the abundance of aquatic insect species, mainly Chironomidae and Simuliidae. Production of ducks was estimated when most young were 2–4 weeks old, either as young per female (Aythya marila and Aythya fuligula) or as total young found (Melanitta nigra, Bucephala islandica, Histrionicus histrionicus). Variation between years in production of young ducks was strongly correlated with variation in the abundance of aquatic insects. Procladius islandicus and Psectrocladius barbimanus, both of which are univoltine with peak emergence in late July (when many young ducks hatch), as well as the bivoltine Orthocladius consobrinus, showed particularly high correlations with duckling numbers. In one year, 1992, catastrophic weather depressed production. Food appeared to determine production, primarily through its availability to the young. Changes in spring density of breeding ducks of all five species were negatively associated with density in the year before. Resources, estimated as food availability or reproductive success in the year before, were positively correlated with change in breeding numbers in A. marila, A. fuligula and B. islandica. Production three years earlier had a weak but significant effect on change in numbers of M. nigra.