Granular Bait Research Articles

Evaluation of metaflumizone granular fly bait for management of houseflies

The housefly, Musca domestica L. (Diptera: Muscidae), is a pest of great veterinary and public health importance. In this study, the efficacy of metaflumizone granular fly bait was assessed on first generation (F1) housefly adults raised from flies collected at a cattle feedlot in Kansas. All bioassays were conducted as choice tests, with flies having ad libitum access to water, granular sugar and bait. A commercial methomyl-based bait (Golden Malrin) was used as positive control; no bait (water and granular sugar only) was used as negative control. Fly mortality was recorded on days 2, 7 and 14. The metaflumizone bait was significantly more slow-acting than the methomyl bait (mortality rates after 2 days of exposure were 49.9% and 57.9%, respectively). However, there were no significant differences in cumulative mortality later in the bioassays. Cumulative mortality rates on days 7 and 14 were 96.1% (metaflumizone), 91.4% (methomyl) and 99.0% (metaflumizone), 97.6% (methomyl), respectively. Our results demonstrate that the metaflumizone granular fly bait may be an effective modality for incorporation into management programmes for houseflies in and around livestock production facilities as well as in residential settings.

Controle de mosca doméstica em área de disposição de resíduos sólidos no Brasil

Desenvolveu-se uma metodologia de controle de Musca domestica (L.) em áreas de disposição de resíduos sólidos. Dois mosquicidas à base de azametifós foram aplicados em diferentes superfícies: nas verticais, a formulação pó molhável foi aplicada com rolos de pintura e, através de pulverização, na superfície de leiras; nas horizontais, foi empregado o mosquicida na formulação granulada. O nível de infestação de moscas foi avaliado por meio do monitoramento em placas (Scudder Fly Grill). Nas áreas das leiras, reduções de 98,5% e 100% foram atingidas em 18 e 30 dias, respectivamente, após a aplicação do produto. Na estação de transferência do lixo, observaram-se reduções de 85,6% e 98,7% no mesmo período de tempo. A aplicação de azametifós em diferentes formulações mostrou ser eficiente no controle da Musca domestica por um período de 30 dias.

Evaluation of Chemical Control Strategies for Linepithema humile (Mayr) (Hymenoptera: Formicidae) in South Carolina State Park Campgrounds1

Abstract The Argentine ant, Linepithema humile (Mayr), is a pest in many urban areas of the southeastern United States. In the Piedmont region of South Carolina, there is a documented problem with L. humile invading the campsites of state park campgrounds. While some parks have tried to implement proactive control programs, the most widely used tactic is spraying insecticides when L. humile populations become intolerable and visitors complain. For this study, a series of four trials were conducted to evaluate insecticidal spray treatments versus combinations of insecticidal sprays and granular bait. Control areas were set up to evaluate the change in ant foraging numbers over a season without chemical interference. One bait (Niban® Granular Bait) and three insecticidal sprays (Premise® 2, Temprid™ SC, and Tempo® Ultra SC) were evaluated to determine the best option in a park setting. The granular bait did not perform as well as the liquid insecticides. Temprid™ SC and Tempo® Ultra SC both resulted in a de...

Formica perpilosa, an Emerging Pest in Vineyards

Formica perpilosa Wheeler is a serious economic ant pest on table grapes grown in the Coachella Valley, California, and Hermosillo, Sonora, Mexico. This ant aggressively tends hemipteran pests, such as the vine mealybug, Planococcus ficus Signoret, and disrupts natural control by predators and parasitoids. Efforts are underway to develop control measures against F. perpilosa using granular bait, yet little is known about the colony life cycle or foraging characteristics of this ant. We studied the seasonal activity, mating behavior, and density and spatial characteristics of F. perpilosa nests in vineyards as well as its foraging and recruitment behavior. Nests were active from early February to mid-October. Mating flights occurred in early August and again in the first two weeks of September and new colonies were founded by a single queen. F. perpilosa rapidly colonized a new, non-infested vineyard with ca. 9% of the vines infested after 1.5 y. In September the proportions of infested vines at 5, 20, and 30 y old vineyards were 18.6, 21.8, and 16.2%, respectively. This ant is seasonally polydomous and nest density increased ca. two-fold at the 5 and 20-year old vineyards between February and September. Foraging and recruitment primarily occurred up to 6.39 m from a home nest. The implications of these studies for controlling F. perpilosa using low-toxic bait delivery systems are discussed.

Relative Attractiveness of Baits to <I>Paratrechina longicornis</I> (Hymenoptera: Formicidae)

Exotic ant incursions are becoming more frequent around the globe, and management with toxic baits is a suitable strategy for most species. Crazy ants, (Latreille) (Hymenoptera: Formicidae), however, are notoriously difficult to attract to commercial baits, which are generally tailored to the preferences of fire ants. We tested P. longicornis preferences for various food types and commercial ant baits. Baits trialed were commercially available products Amdro, Maxforce, Xstinguish (nontoxic monitoring version), Presto, and tuna (in spring water), sugar water (25%), boric acid (1% in 25% sugar water), and deionized water. Tuna and Xstinguish, along with sugar water and sugar water + boric acid, were the most attractive baits to P. longicornis foragers. The granular baits (Maxforce, Amdro, and Presto) were not as attractive to P. longicornis foragers. A decrease in temperature from summer (30 degrees C) to autumn (23 degrees C) trials did not seem to affect the food preferences of P. longicornis. Although P. longicornis recruitment was substantially lower during trials where there was concurrent high native ant abundance and diversity, P. longicornis still recruited to preferred baits in numbers higher than any other species. Given that tuna is impractical for management programs, the effectiveness of boric acid, sweet liquid baits in eliminating P. longicornis colonies should be compared with that of the toxic version of Xstinguish. If both are effective at eliminating colonies, we recommend sweet liquid baits containing boric acid be used for small-scale incursions (one or two nests), but a more practicable solid bait, such as Xstinguish, be used for larger scale incursions (numerous nests).

Relative Attractiveness of Baits to Paratrechina longicornis (Hymenoptera: Formicidae)

Exotic ant incursions are becoming more frequent around the globe, and management with toxic baits is a suitable strategy for most species. Crazy ants, (Latreille) (Hymenoptera: Formicidae), however, are notoriously difficult to attract to commercial baits, which are generally tailored to the preferences of fire ants. We tested P. longicornis preferences for various food types and commercial ant baits. Baits trialed were commercially available products Amdro, Maxforce, Xstinguish (nontoxic monitoring version), Presto, and tuna (in spring water), sugar water (25%), boric acid (1% in 25% sugar water), and deionized water. Tuna and Xstinguish, along with sugar water and sugar water + boric acid, were the most attractive baits to P. longicornis foragers. The granular baits (Maxforce, Amdro, and Presto) were not as attractive to P. longicornis foragers. A decrease in temperature from summer (30°C) to autumn (23°C) trials did not seem to affect the food preferences of P. longicornis. Although P. longicornis recruitment was substantially lower during trials where there was concurrent high native ant abundance and diversity, P. longicornis still recruited to preferred baits in numbers higher than any other species. Given that tuna is impractical for management programs, the effectiveness of boric acid, sweet liquid baits in eliminating P. longicornis colonies should be compared with that of the toxic version of Xstinguish. If both are effective at eliminating colonies, we recommend sweet liquid baits containing boric acid be used for small-scale incursions (one or two nests), but a more practicable solid bait, such as Xstinguish, be used for larger scale incursions (numerous nests).

Evaluation of three formulations of Beauveria bassiana for control of lesser mealworm and hide beetle in Georgia poultry houses.

Initial screening of 12 Beauveria bassiana (Balsamo) Vuillemin isolates against larvae of the lesser mealworm (Alphitobius diaperinus [Panzer]) resulted in the selection of two isolates, GHA and 707, for further testing under field conditions. Three formulations of each strain were prepared: an EC, a ground corn granular formulation, and waste product of fungal propagation containing spent media, mycelia, and unharvested conidia ("residue" formulation). Two field trials were conducted in commercial caged-layer houses in Georgia with 5-6 mo of manure accumulation and established populations of A. diaperinus and hide beetles (Dermestes maculatus DeGeer). In the first trial field, B. bassiana was applied a single time to the manure surface at either 10(9) (EC and granular cornmeal bait formulations) or 10(8) (residue formulation) fungal spores per square meter. In the second trial, two successive weekly treatments were applied, using a total of 6x the rate of application used in the first trial, Significant treatment effects were short-lived and only detected 2 wk after treatment in both trials. The granular formulations of both strains and the residue formulation of the GHA strain provided the greatest degree of suppression (60-90%) of beetle larvae. A laboratory bioassay confirmed that the granular bait was the most effective formulation. More frequent applications made earlier in the manure accumulation cycle may be necessary to achieve satisfactory control of these beetles.

Desenvolvimento de iscas granuladas com atraentes alternativos para Atta bisphaerica Forel, (Hymenoptera: Formicidae) e sua aceitação pelas operárias

Este trabalho teve como objetivos desenvolver uma metodologia para confecção de iscas granuladas a base de atraentes alternativos para as operárias de Atta bisphaerica Forel, e comparar a atratividade destas iscas com a de polpa cítrica. Os carregamentos de grânulos das iscas artesanais de folhas de capim jaraguá e cana-de-açúcar, de colmo de cana-de-açúcar, e de farinha de trigo foram comparados com o da isca comercial sem princípio ativo, em seis formigueiros localizados no município de Viçosa, MG. Concluiu-se que a isca de capim jaraguá foi a preferida pelas operárias de A. bisphaerica, seguida pela isca a base de folhas de cana-de-açúcar. Além da baixa atratividade, verificou-se que as iscas granuladas comerciais apresentaram massa e diâmetro inadequados para o carregamento por essa espécie, dificultando o seu transporte pelas operárias forrageadoras.

Retrieval of granular bait by the Argentine ant (Hymenoptera: Formicidae): effect of clumped versus scattered dispersion patterns.

Argentine ants, Linepithema humile (Mayr), use mass recruitment foraging, with clumped prey items being retrieved more efficiently than dispersed prey. However, in prior field experiments, granular baits, whether dispensed in containers or broadly scattered, had a similar impact on Argentine ant populations. In laboratory experiments, granular insecticide bait was encountered faster by Argentine ant workers and more granules were initially returned to the colony when the granules were scattered versus clumped. After 2 h, granules from both dispersion patterns were retrieved equally. Our results suggest that Argentine ant colonies adjust their foraging patterns to resources of different quality (prey versus bait). Also, foraging activity patterns for bait in the laboratory are consistent with prior field results demonstrating no efficacy advantages to discrete granular bait placements.

Retrieval of Granular Bait by the Argentine Ant (Hymenoptera: Formicidae): Effect of Clumped Versus Scattered Dispersion Patterns

Argentine ants, Linepithema humile (Mayr), use mass recruitment foraging, with clumped prey items being retrieved more efficiently than dispersed prey. However, in prior field experiments, granular baits, whether dispensed in containers or broadly scattered, had a similar impact on Argentine ant populations. In laboratory experiments, granular insecticide bait was encountered faster by Argentine ant workers and more granules were initially returned to the colony when the granules were scattered versus clumped. After 2 h, granules from both dispersion patterns were retrieved equally. Our results suggest that Argentine ant colonies adjust their foraging patterns to resources of different quality (prey versus bait). Also, foraging activity patterns for bait in the laboratory are consistent with prior field results demonstrating no efficacy advantages to discrete granular bait placements.

Development of granular boric acid sugar baits for house flies (Diptera: Muscidae).

Freeze-dried, granular sugar baits containing boric acid as the toxicant were evaluated against house flies, Musca domestica L., in two bioassays with toxicant dose ranges of 3, 5, 7, 8, and 12%, and 3, 5, 9, 17, and 33% boric acid, respectively, calculated and expressed on a dry weight basis. Flies in bioassay 1 were mixed sex and unstarved and flies in bioassay 2 were females only and starved 4 h before testing began. LC50s were 8.97 and 14.33%, and LT50s were 59.75 and 53.34 h in bioassays 1 and 2, respectively. Baits were tested at concentrations as high as 33%, however there was no indication of repellency as seen previously with liquid baits. Reasons for lack of repellency and potential uses for baits are discussed. Development of efficacious granular baits would allow the presentation of higher levels of boric acid in volumes much smaller than required for liquid baits.

MANAGEMENT OF THE STINGING ANT MYRMICA RUBRA, USING A BAITED GRANULAR FORMULATION OF HYDRAMETHYLNON, 1997

Myrmica rubra is an introduced ant that presently occurs on Mount Desert Island, ME, in the vicinity of Acadia National Park, where its painful sting has created problems for residents. Field trials to manage this species with Amdro were conducted at two sites where treated and untreated plots were established: a linear site, with 3.05 by 22.86 m (10 by 75 ft) treated and untreated plots separated by a 3.05 by 30.48-m (10 by 100-ft) buffer plot, and a rectangular site with 18.29 by 27.43-m (60 by 90-ft) treated and untreated plots separated by a 30.48 by 27.43-m (100 by 90-ft) buffer plot. Traps were set at 4.57-m (15-ft) intervals through the middle of the treated, buffer, and untreated plots (total of 16 traps at the linear site and 36 traps at the rectangular site). Ants were trapped using Tic-Tac containers baited with a small piece of cotton soaked with a 50:50 sucrose:water solution set daily for » 2 h between 0800 and 1100 hours. Trapping started on 11 Aug 1997, and a granular ant bait with 0.73% hydramethylnon (Amdro) was applied at a rate of 0.15 kg (1/3 lb) per 929.03 m2 (10,000 ft2) using a seed broadcaster on 18 Aug. Daily trapping continued posttreatment until 26 Sep. Percent control was calculated using a modified Abbott's formula: % control = 100 [(mean in treated area after treatment/mean before)/(mean in untreated area after treatment/mean before) 100]. Changes in ant numbers from the week before to the week after treatment were compared between treated, buffer, and untreated sites by ANOVA using SYSTAT, subprogram MGLH, with data transformed by ln(x+1) to minimize trends in plots of residuals versus estimates.

Management of a Mound‐Building Ant, Lasius neoniger Emery, on Golf Putting Greens and Tees Using Delayed‐Action Baits or Fipronil

The ant Lasius neoniger Emery can be a serious pest when its nesting and mound‐building activities occur on golf putting greens and tees. Controlling turf‐infesting ants is difficult because conventional insecticides usually fail to eliminate the subterranean queen. We compared the attractiveness of commercial and experimental ant baits to L. neoniger workers, and we evaluated baits containing novel, delayed‐action toxicants for selective control of L. neoniger on creeping bentgrass, Agrostis stolonifera L., golf tees. Efficacy of nonbait granular fipronil, a new phenyl pyrazole insecticide, also was evaluated. In choice tests, protein‐based baits were more readily taken than baits containing sugar‐ or oil‐based attractants. Abamectin‐, hydramethylnon‐, or spinosad‐based baits significantly reduced numbers of ant mounds and nests when applied either as spot or broadcast treatments. Two commercial baits, Advance Granular Carpenter Ant Bait, containing 0.011% abamectin B1, and Maxforce Ant Killer Granular Bait, containing 0.9% hydramethylnon, were especially effective. Spot‐treatment with either of these baits, using 1 g of bait per mound, eliminated 88 to 97% of the mounds within 3 d. Broadcast application of fipronil (0.028 kg a.i. ha−1) on heavily infested, sand‐based creeping bentgrass putting green turf in May provided ≈80% reduction of ant mounds at 7 d after treatment and 86% reduction at 60 d.

Observations of fire ants (Solenopsis invicta Buren) attacking apple snails (Pomacea paludosa Say) exposed during dry down conditions

Apple snails (Pomacea paludosa Say) in Florida wetlands and lakes must contend with water level fluctuations that periodically result in dry downs.1 In recent years we have conducted field and laboratory studies to investigate the impacts of such drying events on snail populations. Our research has focused mainly on direct mortality due to desiccation. However, during one laboratory study we identified a potential threat to snail populations from the red imported fire ant, Solenopsis invicta Buren. Since their introduction to Alabama around 1940, S. invicta has expanded its range to include all of Florida.2 S. invicta in Florida prefer disturbed sites in areas with a relatively high water table.3 The highly fragmented wetland systems of Florida provide the moisture levels and disturbance profile (periodic dry downs, as well as disturbed substrata along canals and levees) favored by S. invicta. Fire ants can be found along levees adjacent to parcels of the Everglades and Upper St. Johns River marsh, on tree islands, and along the margins of central Florida lakes (Darby, pers. obs.). Fire ant colonies were reported in up to 20% of alligator nests in central Florida lakes.4 During periods of inundation S. invicta emerge from their mounds and form floating balls, thus prolonging colony survival and aiding dispersal.5 Their adaptations to seasonal flooding and their highly successful invasion of Florida may reflect the evolution of S. invicta in South American ecosystems subjected to regular flooding.6 S. invicta are predators observed killing numerous invertebrate2,6,7 and vertebrate species4,8 and have been associated with biodiversity declines in the southeastern fauna.9 The distribution of S. invicta encompasses the entire range of the Florida apple snail. Given the critical position of the apple snail in the food web of Central and South Florida wetlands,10 vulnerability to S. invicta is of concern. We observed S. invicta attacking apple snails while conducting experiments on snail survival during dry down conditions. Ten tanks (120 cm 3 61 cm 3 46 cm), each with a sand substratum, were set up in a grassy area outside the Florida Caribbean Science Center (USGS-Biological Research Division, Gainesville, FL) in June 1998. S. invicta mounds were present in the grassy area around the tanks. These were treated throughout the experiment with a granular bait formulation of the insecticide hydramethylnon (AMDRO®). [S. invicta identification was verified at the USDA Red Imported Fire Ant Laboratory, Gainesville, FL]. Each tank contained 18–20 adult apple snails with mean (6SD) shell widths of 33.5 mm (63.5 mm). Four tanks were used as controls and kept filled with water, while the other six were Moll. Stud. (1999), 65, 507–510 © The Malacological Society of London 1999

Bait Preference by the Argentine Ant (Hymenoptera: Formicidae) in Haleakala National Park, Hawaii

The Argentine ant, Linepithema humile (Mayr), has proven to be a threat to native arthropod species in Haleakala National Park, Maui, HI, and is also a potential threat to the park's native flora. As it continues to expand its range, an effort has been undertaken to eradicate it, or at the least, control its spread. The 1st part of this effort focused on finding a bait carrier for subsequent toxicant-based control tests. A year-long bait preference test was implemented at each of the ant's 2 infestation sites in Haleakala National Park, in which 6 solid baits and 2 liquid baits were assessed for attractiveness and feasibility for large scale control. At both sites, a toxicant-free formulation of Maxforce, a protein-based granular bait made from ground silkworm, Bombyx mori (L.), pupae, and a 25% sugar water solution were the most attractive baits. Ants took more Maxforce (without toxicant) and sugar water than all other baits, including honey granules and a fish protein bait. Sugar water, however, is difficult to distribute over large natural areas. Maxforce was therefore concluded to be the best bait carrier for toxicant-based control at Haleakala National Park because of its attractiveness and its ease for large scale broadcast dispersal.

Biological Control: An Important Component in Integrated Management ofMusca domestica(Diptera: Muscidae) in Caged-Layer Poultry Houses in Buenos Aires, Argentina

An Integrated Management program to controlMusca domesticaL. was developed by evaluating cultural + chemical control, cultural + biological control + granular bait (IMP 1), and cultural + chemical + biological control + granular bait (IMP 2) in poultry houses in Argentina. Adult fly density was estimated by a modified Scudder-grid method. Percentage of house fly parasitism was estimated by direct and indirect methods.Spalangia endiusWalker andMuscidifurax raptorGirault & Sanders were released at a rate of 10 parasitoids/hen/week. Complete absence of control practices allowed a sustained growth of the fly population. When larvicides and adulticides were used, the number of adult flies decreased but parasitism was adversely affected. When insecticide use was decreased, the number of flies increased. Cultural + biological control + granular bait (IMP 1) produced a reduction of adults from 14 to 3 flies/grid. Where cultural + chemical + biological control + granular bait (IMP 2) was employed, the adult number was reduced from 40 flies/grid to 1 fly/grid. IMP 2 was efficient for reducing the number of larval development sites in the manure. Percentage parasitism exceeded 90% in both IMP 1 and IMP 2 treatments. IMP 1 and IMP 2 treatments both allowed for maximum control of house flies with minimal use of pesticides. With IMP 2, the potential effectiveness of the biological control agents was maximized by reducing adult fly populations with pesticide sprays, then terminating spray applications before parasitoids were released. Thus, IMP 2 would be most suitable for farms which have large adult house fly populations at the initiation of the treatment. The only pesticide used in IMP 1 was a granular bait. Thus, IMP 1 would be most efficient in situations with low initial adult house fly populations and small amounts of manure to ensure the success of the cultural strategy.

Fipronil: An Ultra-Low-Dose Bait Toxicant for Control of Red Imported Fire Ants (Hymenoptera: Formicidae)

Fipronil, a new broad spectrum pyrazole insecticide, was tested both in the laboratory and field as a bait toxicant for control of red imported fire ants, Solenopsis invicta Buren. Laboratory bioassays with worker ants showed that delayed toxicity occurred with baits ranging from 5 to 200 g/ml active ingredient (AI). Tests with field-collected colonies in the laboratory confirmed the bioassay results with worker ants, and demonstrated that granular baits containing from 3.0 to 30 g/mg (AI) eliminated colonies in 8 to 11 weeks after treatment. A field trial showed that a 15 g/mg granular bait provided over 80% colony mortality at 6 and 12 weeks after broadcast application in non-grazed pastures. These results clearly demonstrate the potential of fipronil for use as a bait toxicant for control of red imported fire ants.

Laboratory and Field Evaluation of Granular Ant Baits and The Attractiveness of Liquid Ant Bait, 1996

Abstract Efficacy tests were conducted on two sets of three colonies each of S. invicta and M. pharaonis. An additional colony of each species was held as an untreated control. Each test was replicated three times. Two g of either Ascend PT-370 (modified granular ant bait [.011% Abamectin]) or Maxforce granular ant bait (Hydramethylnon 1 %) were given to each of three colonies in each set, respectively, at the start of the test. Prior to starting the test, all colonies were deprived of food for 48h, to ensure the ants would be hungry. Control colonies were fed the regular laboratory diet . . . live yellow meal-worms for S. invicta and frozen field crickets for M. pharaonois. At the end of 48h bait exposure, bait dishes were removed and reweighed. The amount of bait removed determined the relative attractiveness of the baits. Mortality counts were made beginning 1 wk posttreatment, then weekly for 8 wk. Field tests to determine the attractancy of the experimental ant bait TC-93 to colonies of P. longicornis were conducted on three busy foraging trails. Three active S. invicta mounds were used to test the attractancy of TC-93 to red imported fire ants. These tests were conducted by placing lg of TC-93 liquid ant bait in the center of 4 X 5 inch note cards, and placing a baited card on each of 3 busy P. longicornis forging trails. At TC-93 baited card was also placed at the bases of 3 active S. invicta mounds. Counts of ants visiting these stations were made over a 2-min period at 10, 30, and 60 min. following bait placement. The number of ants responding determined the relative attractiveness to the ants. Tests were conducted at Moody Gardens, Galveston, TX to determine the relative attractiveness of the TC-93 ant bait to Ghost ants, and to compare the measured response to that stimulated by the granular ant baits Maxforce and Ascend PT-370 modified bait. These tests utilized 24 bait stations: 8 stations contained 1 g each TC-93; 8 contained 1 g each Maxforce; 8 contained 1 g each Ascend PT-370 (modified) ant bait. These stations were clear, covered plastic petri dishes (35mm X 10mm) with lids removed to accommodate ant entry. These stations were alternately placed about lm apart in areas where ants were seen. Ghost ants (and other species) were counted at the bait stations 30 min after placement. A final count was made 2h post-treatment. Bait stations were then picked up and returned to the lab and reweighed to determine the amount of bait removed to assess and compare the relative attractiveness of each bait.

Field Evaluation of Perimeter Treatments for Pharaoh Ant (Hymenoptera: Formicidae) Control

Four types of exterior perimeter treatments for the control of Pharaoh ants, Monomorium pharaonis (L.), in houses were compared. Treatment with a granular bait containing hydramethylnon, a delayed-action toxicant, resulted in a 95% reduction in foraging activity 1 and 4 weeks after treatment. Houses treated with either an insecticidal residual spray, insecticide granules, or a granular bait containing a contact insecticide, had an average increase of 22% in foraging activity. Foraging by Pharaoh ants on a delayed-action toxicant bait applied solely to the exterior periphery of houses controlled Pharaoh ant colonies located inside the structure. Exterior perimeter baiting for Pharaoh ants could reduce indoor insecticide applications.

Control of the behaviour of leaf-cutting ants by their ‘symbiotic’ fungus

There is an obligatory relationship between leaf-cutting ants of the generaAtta andAcromyrmex (Hymenoptera; Attini) and the fungus,Attamyces bromatificus Kreisel, for which they provide a substrate of cut plant material. We show that the ants learn to reject plant material that contains chemicals injurious to the fungus. After an initial period of acceptance, ants from laboratory nests stopped harvesting granular bait containing a fungicidal agent (cycloheximide) and orange peel. This rejection was maintained for many weeks. These colonies also rejected control bait containing no cycloheximide. Some generalisation of the response was observed; colonies rejecting orange granules also rejected grapefruit granules, although they still accepted blackcurrant granules. Rejection of fungicidal bait by colonies in the field was restricted to ants on foraging trails exposed to experimental bait. We conclude that a semiochemical from the fungus, circulated by trophallaxis and grooming, regulates the selection of plant material by foragers. The fungus gains more advantages from the symbiotic relationship than is often realised. The ants provide the fungus with housing, sanitation, defence against disease and predators, pre-selected food, and a means of dissemination, at the cost to itself of providing food for the ant brood.