Experimental evidence for the costs and hygienic significance of the antibiotic metapleural gland secretion in leaf-cutting ants

The paired exocrine metapleural glands present in the large majority of ant species produce compounds with antibiotic properties. In the leaf-cutting ant, Acromyrmex octospinosus, the secretion consists of more than 20 different compounds and it has generally been assumed that the glands serve as a general defence against various infectious microbes of fungal and bacterial origin. We present results illuminating the direct costs and benefits of these metapleural gland defences in A. octospinosus. We show that major workers of this leaf-cutting ant experience a significant reduction in their respiration rate when the metapleural glands are experimentally closed, indicating that metapleural gland secretion incurs a substantial cost and that the production of compounds from these glands is terminated when the ants are incapable of secreting them. In another set of experiments, we show that the ability to secrete antibiotic compounds from the metapleural glands is of significant importance when ants are exposed to a general but potentially virulent insect pathogen, Metarhizium anisopliae. Infection with this fungus is lethal within a few days when ants have their metapleural glands experimentally closed, but relatively harmless when the metapleural glands are functional. These findings support experimentally the view that the metapleural glands play an important hygienic role in leaf-cutting ants.     

Evidence for differential selection and potential adaptive evolution in the worker caste of an inquiline social parasite

Social parasites exploit the socially managed resources of social insect colonies in order to maximise their own fitness. The inquilines are among the most specialised social parasites, because they are dependent on being fully integrated into their host's colony throughout their lives. They are usually relatives of their host and so share ancestral characteristics (Emery's rule). Closely related inquiline-host combinations offer a rare opportunity to study trade-offs in natural selection. This is because ancestral adaptations to a free-living state (e.g. the production of a worker caste) become redundant and may be replaced by novel, parasitic traits as the inquiline becomes more specialised. The dynamics of such processes are, however, unknown as virtually all extant inquiline social parasites have completely lost their worker caste. An exception is Acromyrmex insinuator, an incipient permanent social parasite of the leaf-cutting ant Acromyrmex echinatior. In the present study, we document the size distribution of parasite and host workers and infer how selection has acted on A. insinuator to reduce, but not eliminate, its investment in a worker caste. We show that the antibiotic producing metapleural glands of these parasite workers are significantly smaller than in their host counterparts and we deduce that the metapleural gland size in the host represents the ancestral state. We further show experimentally that social parasite workers are more vulnerable to the general insect pathogenic fungus Metarhizium than are host workers. Our findings suggest that costly disease resistance mechanisms are likely to have been lost early in inquiline evolution, possibly because active selection for maintaining these traits became less when parasite workers had evolved the ability to exploit the collective immune system of their host societies.     

Sanitizing the fortress: protection of ant brood and nest material by worker antibiotics

Social groups are at particular risk for parasite infection, which is heightened in eusocial insects by the low genetic diversity of individuals within a colony. To combat this, adult ants have evolved a suite of defenses to protect each other, including the production of antimicrobial secretions. However, it is the brood in a colony that are most vulnerable to parasites because their individual defenses are limited, and the nest material in which ants live is also likely to be prone to colonization by potential parasites. Here, we investigate in two ant species whether adult workers use their antimicrobial secretions not only to protect each other but also to sanitize the vulnerable brood and nest material. We find that, in both leaf-cutting ants and weaver ants, the survival of the brood was reduced and the sporulation of parasitic fungi from them increased, when the workers nursing them lacked functional antimicrobial-producing glands. This was the case for both larvae that were experimentally treated with a fungal parasite (Metarhizium) and control larvae which developed infections of an opportunistic fungal parasite (Aspergillus). Similarly, fungi were more likely to grow on the nest material of both ant species if the glands of attending workers were blocked. The results show that the defense of brood and sanitization of nest material are important functions of the antimicrobial secretions of adult ants and that ubiquitous, opportunistic fungi may be a more important driver of the evolution of these defenses than rarer, specialist parasites.     

Waste management in the leaf-cutting ant Acromyrmex echinatior: the role of worker size, age and plasticity

Division of labour is the hallmark of the success of many social animals. It may be especially important with regard to waste management because waste often contains pathogens or hazardous toxins and worker specialisation can reduce the number of group members exposed to it. Here we examine waste management in a fungus-farming, leaf-cutting ant, Acromyrmex echinatior, in which waste management is necessary to protect their vulnerable fungal crop. By marking ants with task-specific paint colours, we found clear division of labour between workers that engage in waste management and those that forage, at least during the fine timescale of the 3-day marking period. This division of labour was influenced by both age and size, with waste management workers tending to be smaller and younger than foragers. The role of preventing contaminated ants from entering the colony was fulfilled mainly by medium-sized workers. When the level of waste was experimentally increased, most of the ants that responded to remove the waste were workers previously engaged in tasks inside the nest rather than external waste workers or foragers. These responding workers tended to be young and medium-sized. Surprisingly, the responding ants were subsequently able to revert back to working within the fungus garden, but the probability of them doing so depended on their age and the length of time they were exposed to waste. The results demonstrate the importance of division of labour with regard to waste management in A. echinatior and show that this is adaptable to changing needs.     

Defense against parasites by hitchhikers in leaf-cutting ants: a quantitative assessment

Summary A curious behavior in leaf-cutting ants in the genus Atta is the hitchhiking of small minim workers on leaf fragments carried by larger workers. Two functions of these hitchhikers have been proposed: (1) defense of leaf carriers against parasitic flies in the family Phoridae (ant protection hypothesis; Eibl-Eibesfeldt and Eibl-Eibesfeldt 1967) and (2) reduction of transport costs of small minims that collect plant fluids (energy conservation hypothesis; Stradling 1978).We studied hitchhiking behavior in colonies of Atta colombica on Barro Colorado Island, Panama, and found strong evidence in favor of the ant protection hypothesis. Females of Apocephalus attophilus (Diptera: Phoridae) attack leaf carriers of A. colombica and deposit eggs in the head capsules of these ants. Our observations indicate that parasites require leaf fragments to stand on during oviposition, and, as a result, only leaf carriers are susceptible to parasitic attack. The presence of hitchhikers reduces significantly both the time parasites spend on leaf fragments and the probability that they will land in the first place. Results of experimental introductions of parasites and a year of biweekly censuses at ten colonies indicate that leaf-cutting ants adjust the level of hitchhiking to accommodate both daily and seasonal changes in the abundance of parasites.We found little evidence in support of the energy conservation hypothesis. If it is assumed that all minim workers hitchhike back to the nest, our calculations indicate that total transport costs along a foraging trail are reduced by 10% or less. However, our observations indicate that only 50% of returning minim workers hitchhike, and therefore energy savings are actually considerably less than 10%. Leaf-cutting ants in the genera Atta and Acromyrmex are attacked by over 20 species of parasitic phorids. In the discussion we review what is known about these associations and suggest that these parasites have influenced the ecology and evolution of polyethism in leaf-cutting ants. Offprint requests to: D.H. Feener (at his present address)     

Predatory behavior and chemical communication in two Metapone species (Hymenoptera:Formicidae)

Summary. Colonies of two species of Metapone (M. madagascarica, M. new species.) were collected in Madagascar and established in laboratory nests. It could be demonstrated that both species are specialist predators of termites (Cryptotermes kirbyi). During hunting the ants sting the termites and thereby paralyze and preserve the prey alive. In this way prey can be stored in the ant nest for extended periods. During foraging and colony emigrations the ants lay chemical trails with poison gland secretions. Among the seven compounds identified in the venom only methyl pyrrole-2-carboxylate elicits trail following behavior in both Metapone species. Received 11 February 2002, accepted 23 February 2002.     

The origin of the chemical profiles of fungal symbionts and their significance for nestmate recognition in <Emphasis Type="Italic">Acromyrmex</Emphasis> leaf-cutting ants

Cuticular hydrocarbon profiles are essential for nestmate recognition in insect societies, and quantitative variation in these recognition cues is both environmentally and genetically determined. Environmental cues are normally derived from food or nest material, but an exceptional situation may exist in the fungus-growing ants where the symbiotic fungus garden may be an independent source of recognition compounds. To investigate this hypothesis, we quantified the chemical profiles of the fungal symbionts of 18 sympatric colonies of Acromyrmex echinatior and Acromyrmex octospinosus and evaluated the quantitative variation of the 47 compounds in a multivariate analysis. Colony-specific chemical profiles of fungal symbionts were highly distinct and significantly different between the two ant species. We also estimated the relative genetic distances between the fungal symbionts using amplified fragment length polymorphism (AFLP) and correlated these with the overall (Mahalanobis) chemical distances between the colony-specific profiles. Despite the standardized laboratory conditions, the correlations were generally weak, but a statistically significant portion of the total variation in chemical profiles could be explained by genetic differences between the fungal symbionts. However, there was no significant effect of ant species in partial analyses because genetic differences between symbionts tend to coincide with being reared by different ant species. However, compound groups differed significantly with amides, aldehydes, and methyl esters contributing to the correlations, but acetates, alkanes, and formates being unrelated to genetic variation among symbionts. We show experimentally that workers that are previously exposed to and fed with the fungal symbiont of another colony are met with less aggression when they are later introduced into that colony. It appears, therefore, that fungus gardens are an independent and significant source of chemical compounds, potentially contributing a richer and more abundant blend of recognition cues to the colony “gestalt” than the innate chemical profile of the ants alone. Freddie-Jeanne Richard and Michael Poulsen contributed equally to this work.     

Chemical trail communication in the amblyoponine species Mystrium rogeri Forel (Hymenoptera, Formicidae, Ponerinae)

Summary. Workers of the amblyoponine species Mystrium rogeri employ trail communication during recruitment to food sources and new nest sites. The trail pheromone originates from a hitherto unknown sternal gland located in the 7th abdominal sternite. The recruiting ant deposits the gland secretions by a special gaster-dragging behavior. The recruitment behavior can be complemented by a rapid vertical body shaking performed by some recruiting ants inside the nest. M. rogeri workers possess a large pygidial gland, the secretion of which elicits a repellent response in other ant species. Received 25 May 1998; accepted 15 June 1998.     

Three beetles—three concepts. Different defensive strategies of congeneric myrmecophilous beetles

Myrmecophiles, i.e., organisms associated with ants live in a variety of ecological niches in the vicinity or inside ant colonies and employ different strategies to survive ant encounters. Because different niches are characterized by different encounter rates with host ants, strategies used to avoid ant aggressions should depend on these niches. This hypothesis was studied with three rove beetle species of the genus Pella, which are myrmecophiles of the ant Lasius fuliginosus and the non-myrmecophilous relative Drusilla canaliculata. Behavioral tests in the field revealed that Pella species are better adapted to interactions with ants than D. canaliculata, but that they use species-specific strategies in ant interactions. Pella cognata and Pella funesta avoid encounters with ants by swift movements. Chemical analyses of the defensive tergal gland secretions showed that P. cognata has replaced the aggression inducing undecane by the behaviorally neutral tridecane. P. funesta repels the ants by releasing the panic alarm pheromone sulcatone from its tergal gland resulting in an “ant free space” around the beetles. Finally, Pella laticollis uses a specific and unique appeasing behavior. Behavioral and chemical data did not reveal any indication for the mimicry of the ants' cuticular hydrocarbon profiles by any of the beetle species. It is discussed that the employed strategies correlate with the ecological niches of the beetles. P. cognata and P. funesta are living along ant trails with ample space to escape and the employed strategies are probably sufficient to escape from dangerous conflicts. In contrast, P. laticollis lives in refuse areas of ant nests with frequent ant encounters, and its appeasement strategy allows it to stay at the encounter site.     

Allometry and the geometry of leaf-cutting in Atta cephalotes

Summary This study considers the relationship of both leg length and the geometry of leaf-cutting to load-size determination by the highly polymorphic leaf-cutting ant Atta cephalotes. A. cephalotes workers anchor on the leaf edge by their hind legs and pivot around them while cutting arcs from leaves. I tested the hypothesis that, for an ant cutting a semicircular leaf fragment, fragment area is determined by a fixed reach while cutting. This reach hypothesis predicts that ants should cut the same fragment-area for at all leaf types. Also, if the radius of the semicircular fragment is proportional to hind leg length, this hypothesis predicts that leaf area should be proportional to hind-leg length squared. The field work was carried out in March–April 1990 and June 1991 in Heredia Province, Costa Rica. I measured hind-leg length for workers of different masses. I then measured leaf-fragment area and mass for workers cutting semicircular fragments from leaves of different densities (mass/area). The logarithmic relationship between ant mass (M _a) and hind-leg length L accelerated negatively (Fig. 1). As a result of this complex allometry, relative leg length (L/M _a ^0.33) increased with ant mass up to a mass of 7.4 mg. Above 7.4 mg, relative leg length decreased. For foragers cutting semicircular fragments, the area cut by an ant of a given size showed no significant difference among leaves of different densities (Fig. 2). Leaf area (A) increased as a function of leg length to the 1.9 power (Fig. 4), an exponent not significantly different from the square function expected if the radius of a fragment is determined by the ant's reach. As a result of this consistent mode of fragment-area determination, the mass of fragments cut by an ant of a given size was significantly greater when cutting denser leaves (Fig. 3) and relative area (A/M _a) cut decreased with increasing ant mass. However, because larger ants generally cut denser leaves (Table 1), the increased density of thick leaves was offset by the reduced relative area cut by the larger ants. Overall, 93% of the foragers cut fragments weighing between 1.5 and 6 times their own body mass (Table 1). Earlier studies found that this broad load-mass range maximized the biomass-transport rate (mass/distance/time) and transport efficiency (mass/distance/energy cost). Thus, A. cephalotes does not solve the problem of matching ant mass and load mass at leaves of different densities with flexibility in the leaf-cutting behavior of individual ants. Instead, individual ants employ a single simple behavioral rule, but workers of different sizes and body proportions tend to cut leaves of different densities.     

Both evaluation of resource quality and speed of recruited leaf-cutting ants (Acromyrmex lundi) depend on their motivational state

Summary During recruitment, running velocity of both outbound and laden workers of the leaf-cutting ant Acromyrmex lundi depended on the information about resource quality they received from the first successful recruiter. In independent assays, single scout ants were allowed to collect sugar solutions of different concentrations and to recruit nestmates. Recruited workers were presented with standardized paper discs rather than the sugar solution given to the original recruiting ant. Outbound recruited workers were observed to run faster the more concentrated the solution found by the recruiter. Speed of disc-laden workers also depended on the concentration of the solution found by the recruiter, i.e. on the information about food quality they received, since they had no actual contact with the sugar solution. Disc-laden workers ran, as intuitively expected, slower than outbound workers. The reduction in speed, however, could not be attributed to the effects of the load itself, because workers collecting discs of the same weight, but with added sugar, ran as rapidly as outbound, unladen workers. Workers collecting standardized sugared discs reinforced the chemical trail on their way to the nest. The percentage of trail-layers was higher when workers were recruited to 10% than to 1% sugar solution, even though they collected the same kind of discs at the source. Their evaluation of resource quality, therefore, depended on their motivational state, which was modulated by the information they received during recruitment. Using previously published data on energetics of locomotion in leaf-cutting ants, travel costs of A. lundi workers recruited to sugar solutions of different concentration could be estimated. For workers recruited to the more concentrated solution, both speed and oxygen consumption rate increased by a roughly similar factor. Therefore, although workers ran faster to the high-quality resource, their actual energy investment per trip remained similar to that made by workers recruited to the low-quality resource. It is suggested that the more motivated workers reduced travel time without increasing energy costs during the trip. The adaptive value of these responses seems to be related to a rapid transmission of information about a newly discovered food source.     

Foaming in Pachycondyla: A new defense mechanism in ants

Summary When disturbed, two species of Malayan Pachycondyla release foam threads more than 10 cm in length or foam piles. The source of the proteinaceous foam is the enlarged venom gland, which is probably frothed up by air from the spiracles of the spiracular plates. The Dufour's gland normally producing hydrocarbons in stinging ants is atrophied. Therefore, absence of the Dufour's gland could be essential to the foaming ability, since the lipophilic hydrocarbons inhibit froth production in protein solutions. The release of foam is a mechanically acting defense mechanism, which is very effective against small mass-attacking ants. Pachycondyla species are also able to sting effectively.     

Induced responses to herbivory in the Neotropical ant-plant association between Azteca ants and Cecropia trees: response of ants to potential inducing cues

Plant defense against herbivores often involves constitutive and inducible mechanisms of resistance. Obligate ant-plants, which provide food and housing for ants, are thought to primarily rely on ants for defense against herbivores. This form of plant defense has largely been viewed as static. We have been investigating the dynamic nature of Azteca ants as an inducible defense of Cecropia trees. Ants rapidly recruit to and patrol sites of foliar damage. We propose that Azteca ants can be viewed as an inducible defense for Cecropia trees because of their sensitivity to cues associated with herbivory, their rapid and aggressive recruiting ability, and their reclaimable and redeployable nature as a plant defense. In this study, we examine ant behavior following plant damage, and the potential cues that indude ant recruitment. We found that ants present on leaves when the plant is damaged leave the damaged leaf and recruit other ants to it, presumably by laying recruitment trails. Volatile leaf cues associated with herbivory were important in eliciting an induced response in two experiments. However, we found that cues associated with a congeneric plant elicited a much stronger ant response than conspecific cues. Although the type of leaf damage (gaping wounds versus leaf edge wounds) did not affect the level of ant recruitment, the extent of damage did. Leaves with one hole punched showed a 50% increase in ants, while leaves with five holes punched in them elicited a 100% increase in ant numbers. In sum, it appears that multiple plant-related cues associated with herbivory are involved in induction of ant recruitment in the Cecropia-Azteca system. We discuss the generality of ant responses to herbivory in obligate ant-plant systems, and in facultative ant-plant associations, which may be more common. Received: 23 March 1998 / Accepted after revision: 5 July 1998     

Fight or flight? A geographic mosaic in host reaction and potency of a chemical weapon in the social parasite <Emphasis Type="Italic">Harpagoxenus sublaevis</Emphasis>

Ant social parasites use chemical warfare to facilitate host colony takeover, which is a critical but recurring step in their life cycle. Many slave-making ants use the secretion of the Dufour gland to manipulate host behaviour during parasitic nest foundation and slave raids. Harpagoxenus sublaevis applies this chemical weapon onto defending Leptothorax host workers, which elicits deadly fights amongst them. Host species are expected to evolve counter-adaptations against this behavioural manipulation and in this study we investigated the geographic structure of this co-evolving trait. We compared the effectiveness of the parasitic gland secretion from different H. sublaevis populations in host colonies from various sites and analysed the occurrence of local adaptation. The two host species L. muscorum and L. acervorum generally showed different responses to the parasites’ chemical weapon: L. acervorum attacked nestmates treated with Dufour gland secretion, while L. muscorum workers fled. Flight, instead of intraspecific fights, is an adaptive host reaction as it results in fewer host fatalities during raids. Beside interspecific host differences, we found a geographic mosaic of host resistance: parasites from a German population strongly manipulated the behaviour of both sympatric Leptothorax populations. Russian or Italian hosts instead did not react with intracolonial aggression, but fled when confronted with the gland secretion of their sympatric parasite. Not only variation in host resistance explains differences in the effectiveness of the parasitic gland secretion but also interpopulational differences in its chemical composition, which were revealed by gas chromatography and mass spectrometry.     

Parasitoid learning during interactions with ants: how to deal with an aggressive antagonist

Pauesia picta, P. pinicollis and P. silvestris (Hymenoptera: Aphidiidae) are common parasitoids of the conifer aphid Cinara pinea, which is regularly attended by red wood ants (Formica spp.). In this study, I tested whether females of these parasitoid species learned during interactions with honeydew-collecting Formica polyctena workers that caution is not necessary if searching behaviour is adapted, and whether parasitoids benefit from being able to learn. When searching on Scots pines, naive females of P. picta and P. pinicollis generally retreated to a pine needle when making contact with a honeydew-collecting ant, did not approach ants from the side or from the front and kept a ”safe distance” from ants when sitting on needles. After some non- aggressive ant encounters, experienced female parasitoids changed their behaviour: they reduced their searching speed, approached ants from the side and even from the front, retreated less often in response to an approaching ant and reduced the ”safe distance”. These experienced females had a significantly higher rate of oviposition than naive females or females foraging for an unattended host. Thus, the ability of the parasitoid to learn during interactions with an antagonist led to a prolonged retention time and a higher oviposition rate. By contrast, there was no evidence of learning in P. silvestris. Females of this species showed no behavioural change in response to ant encounters, and there was no difference in the foraging success of naive and experienced female parasitoids. Received: 7 December 1999 / Revised: 23 September 2000 / Accepted: 10 March 2000     

Multiple mating and facultative polygyny in the Panamanian leafcutter ant Acromyrmex echinatior

Queen mating frequency of the facultatively polygynous ant Acromyrmex echinatior was investigated by analysing genetic variation at an (AG)_n repeat microsatellite locus in workers and sexuals of 20 colonies from a single Panamanian population. Thirteen colonies were found to be monogynous, 5 colonies contained multiple queens, whereas the queen number of 2 colonies remained unresolved. Microsatellite genotypes indicated that 12 out of 13 queens were inseminated by multiple males (polyandry). The mean queen mating frequency was 2.53 and the mean genetically effective paternity frequency was 2.23. These values range among the highest found in ants, and the results are in keeping with the high mating frequencies reported for other species of leafcutter ants. Consistent skew in the proportional representation of different patrilines within colonies was found, and this remained constant in two consecutive samples of offspring. Dissections showed that all examined queens from multiple-queen colonies were mated egg-layers. The mean relatedness value among nestmate workers in polygynous colonies was lower than that for monogynous colonies. No diploid males were detected in a sample of 70 genotyped males. Worker production of males was detected in one queenless colony. We discuss our findings in relation to known patterns of multiple maternity and paternity in other eusocial Hymenoptera. Received: 2 September 1998 / Received in revised form: 3 February 1999 / Accepted: 7 February 1999     

Toxicity of the venom in three neotropical Crematogaster ants (Formicidae: Myrmicinae)

Bioassays were performed to investigate and compare the toxic properties of the contact venoms of three Neotropical Crematogaster species (C. sp. prox. abstinens, C. distans and C. brevispinosa rochai), whose main venom compounds are chemically different (long chain derivatives linked to an aldehyde or a primary acetate, and furanocembrenoid diterpenes, respectively). Different quantities of venom were topically applied on the bodies of three target insect species (workers of the termite Nasutitermes ephratae, workers of the ant Solenopsis sp., “media” and “major” workers of the ant Cephalotes pusillus). The toxicity of the venom greatly varied according to the Crematogaster species, the venom of C. b. rochai showing the highest toxic activity towards all target species. The sensitivity of the target species also greatly varied, the workers of N. ephratae showing a great sensitivity to the venoms, while the heavily armoured workers of the ant C. pusillus, particularly “major” ones, were resistant. Cuticle thickness was found to be a key factor in the resistance to the toxic effects of Crematogaster venom. Sensitivity of the target workers was also directly related to the amount of venom applied to their cuticle. Workers of C. distans were resistant to topical application of their own venom, while C. sp. prox. abstinens and C. b. rochai were not. In all cases, the venoms were slow acting, as several hours to one or even several days were needed to obtain a significant mortality. The main biological activities of the Crematogaster venom could be related to more immediate repellent properties, as recently shown with the European species C. scutellaris.     

Global optimization from suboptimal parts: foraging sensu lato by leaf-cutting ants

Central-place foraging theory has been unable to explain the load selection behavior of leaf-cutting ants (Atta spp., Attini: Formicidae). We suggest that this is due to incomplete consideration of the sequence of behaviors involved in resource acquisition by these ants. Unlike most central-place foragers, leaf-cutting ants do not return to their nests with food. Instead, the leaf fragments they gather must be processed within the nest to convert them to substrate for fungal gardens. We have shown previously that leaf fragment size affects the rate of distribution and processing of leaf tissue inside laboratory nests of Atta colombica. Including these tasks in the calculation of foraging rate may help explain load selection and other features of central-place foraging by Atta colonies. Here we develop a mathematical model of the complete sequence of external and internal tasks that lead to addition of substrate to fungal gardens. Using realistic parameter values, the leaf fragment sizes predicted to maximize a colony's rate of foraging in this broad sense correspond well with the mean fragment sizes actually collected by Atta colonies in the field. The optimal fragment size for global performance in the model is below the size that would maximize the delivery rate by above-ground foragers. The globally optimal size also fails to maximize the rate of either fragment distribution or fragment processing within the nest. Our results show how maximum collective performance of an ensemble of linked tasks may require behavior that would appear suboptimal in a piecemeal analysis of tasks.     

The costs and benefits of cooperation between the Australian lycaenid butterfly,Jalmenus evagoras,and its attendant ants

Summary The larvae and pupae of the Australian lycaenid butterfly, Jalmenus evagoras associate mutualistically with ants in the genus Iridomyrmex. Four ant exclusion experiments in three field sites demonstrated that predation and parasitism of J. evagoras are so intense that individuals deprived of their attendant ants are unlikely to survive. Larvae and pupae of J. evagoras aggregate, and the mean number of attendant ants per individual increases with larval age and decreases with group size. Field observations showed that young larvae could gain more attendant ants per individual by joining the average size group of about 4 larvae than by foraging alone. Aggregation behaviour is influenced by ant attendance: young larvae and pupating fifth instars aggregated significantly more often on plants with ants than on plants where ants had been excluded. In return for tending and protecting the larvae, ants were rewarded by food secretions that can amount to as much as 409 mg dry biomass from a single host plant containing 62 larvae and pupae of J. evagoras over a 24 h period. Larval development in the laboratory lasted approximately a month, and larvae that were tended by ants developed almost 5 days faster than larvae that were not tended. However, tended individuals, particularly females, pupated at a significantly lower weight than their untended counterparts, and the adults that eclosed from these pupae were also lighter and smaller. On average, pupae that were tended by ants lost 25% more weight than untended pupae, and in contrast with larvae, they took longer to eclose than pupae that were not tended. These experimental results are discussed in terms of costs and benefits of association for both partners, and of aggregation for the lycaenids.     

Learning,retention and coding of nest-associated visual cues by the Australian desert ant, <Emphasis Type="Italic">Melophorus bagoti</Emphasis>

A variety of social insects use visual cues for homing. In this study, we examine the possible factors affecting the learning and retention of nest-associated visual cues by the Australian desert ant Melophorus bagoti and the manner in which such cues are encoded by foraging ants. We placed four prominent cylindrical landmarks around a nest and trained foragers from that nest to a food source. Ants were tested with the landmark array in a distant testing field after (1) a known number of exposures to the landmarks (1, 3, 7 or 15 trials, spread over a period of 1 day, 2 days or ≥3 days) and (2) after a known period of delay (0, 24, 48, 96 or 192 h). The results show that a combination of an increase in training trials and an increase in number of training days affected the acquisition of landmark memory. Moreover, once the landmarks were learnt, they became a part of long-term memory and lasted throughout the ants’ foraging lifetime. To examine visual cue encoding behaviour, ants trained under similar conditions for 4 days were tested with (1) an identical landmark array, (2) landmarks of the same size used in training, but placed at twice the distance from each other, and (3) landmarks whose dimensions were doubled and placed at twice the distance from each other. In conditions (1) and (3), the ants searched extensively at the centre of the four landmarks, suggesting that, similar to the Saharan ant (genus Cataglyphis) and the honeybee, M. bagoti too uses a snapshot to match the view of the landmarks around the nest. But contrary to the snapshot model, in condition (2), the ants did not search extensively at the centre of the landmarks, but searched primarily 0.5 m from the landmark, the distance from each landmark to the nest during training. We discuss how various search models fare in accounting for these findings.