Body size is a universal property affecting biological structure and function, from cell metabolism to animal behavior. The nervous system, the physical generator of behavior, is also affected by variations in body size; hence potentially affecting the way animals perceive, interpret and react to the environment. When animals join to form groups, such individual differences become part of the structure of the society, even determining social roles. Here, I explore the association between body size, behavior and social organization in honeybees and bumblebees. Focusing on bumblebees, I explore the link between body size, brain allometry and learning and memory performance, within the context of task specialization. I show that body size goes along with brain size and with learning and memory performance, and that foraging experience affects such cognitive and neural features. Next, I explore the association between body size and foraging task specialization in honeybees. Previous evidence showed a link between specialization on pollen or nectar foraging and sensory sensitivity, further associating sensitivity to the quality and/or quantity of resource exploited. I hypothesize that, as in solitary bees, larger body size is associated with higher sensory sensitivity. I test this hypothesis by comparing body size and the quality and quantity of the resource exploited by wild Africanized and European honeybees. I show that nectar foragers are smaller and have fewer olfactory sensilla, which might underlie their lower sensitivity to odors. Also, larger bees collect more pollen (within pollen foragers) and more dilute nectar (within nectar foragers). To further test this `size hypothesis', I compare strains of bees selected to store large ("high strain") or small ("low strain") amounts of pollen surplus. As these strains differ in sensory sensitivity, I predict that the more sensitive high strain bees are larger and have more sensory sensilla. I show that high strain bees are generally bigger, but have fewer sensory sensilla than low strain bees. These results show that in bees, body size is associated with an individual's sensory, neural and cognitive features, further suggesting that body size plays a more important role in the organization of bee societies than generally assumed.

Thesis (MSc.(Chemistry)) - University of Pretoria, 2008. / Includes bibliographical references

Ants are fascinating creatures - not so much because they are intelligent on their own, but because as a group they display compelling emergent behaviour (the extent to which one observes features in a swarm which cannot be traced back to the actions of swarm members). What does each swarm member do which allows deliberate engineering of emergent behaviour? We investigate the development of a language for programming swarms of ant agents towards desired emergent behaviour. Five aspects of stigmergic (pheromone sensitive computational devices in which a non-symbolic form of communication that is indirectly mediated via the environment arises) and message passing ant agents (computational devices which rely on implicit communication spaces in which direction vectors are shared one-on-one) are studied. First, we investigate the primitive behaviours which characterize ant agents' discrete actions at individual levels. Ten such primitive behaviours are identified as candidate building blocks of the ant agent language sought. We then study mechanisms in which primitive behaviours are put together into XSets (collection of primitive behaviours, parameter values, and meta information which spells out how and when primitive behaviours are used). Various permutations of XSets are possible which define the search space for best performer XSets for particular tasks. Genetic programming principles are proposed as a search strategy for best performer XSets that would allow particular emergent behaviour to occur. XSets in the search space are evolved over various genetic generations and tested for abilities to allow path finding (as proof of concept). XSets are ranked according to the indices of merit (fitness measures which indicate how well XSets allow particular emergent behaviour to occur) they achieve. Best performer XSets for the path finding task are identifed and reported. We validate the results yield when best performer XSets are used with regard to normality, correlation, similarities in variation, and similarities between mean performances over time. Commonly, the simulation results yield pass most statistical tests. The last aspect we study is the application of best performer XSets to different problem tasks. Five experiments are administered in this regard. The first experiment assesses XSets' abilities to allow multiple targets location (ant agents' abilities to locate continuous regions of targets), and found out that best performer XSets are problem independent. However both categories of XSets are sensitive to changes in agent density. We test the influences of individual primitive behaviours and the effects of the sequences of primitive behaviours to the indices of merit of XSets and found out that most primitive behaviours are indispensable, especially when specific sequences are prescribed. The effects of pheromone dissipation to the indices of merit of stigmergic XSets are also scrutinized. Precisely, dissipation is not causal. Rather, it enhances convergence. Overall, this work successfully identify the discrete primitive behaviours of stigmergic and message passing ant-like devices. It successfully put these primitive behaviours together into XSets which characterize a language for programming ant-like devices towards desired emergent behaviour. This XSets approach is a new ant language representation with which a wider domain of emergent tasks can be resolved.

Made available in DSpace on 2016-03-07T19:20:42Z (GMT). No. of bitstreams: 0 Previous issue date: 2015-09-10. Added 1 bitstream(s) on 2016-03-07T19:24:21Z : No. of bitstreams: 1 000858914.pdf: 3359407 bytes, checksum: a2fdb848e700d9d5c100c7ff4aeb7c8e (MD5) / Entre todas as savanas do mundo, o Cerrado tem a maior biodiversidade e tem sido a mais ameaçada pela conversão de terras. Os impactos resultantes de mudanças de uso da terra sobre a biodiversidade não foram avaliados e a restauração do Cerrado é um desafio ainda a ser superado. Nós investigamos ambos - as perdas de diversidade de plantas e formigas devido à conversão da terra que compreende sistemas de produção (cana-de-açúcar, pastagem, silvicultura de ciclo curto e ciclo longo) e o sucesso de diferentes técnicas de restauração (passiva ou ativa, esta envolvendo o plantio de mudas em povoamentos puros e mistos), utilizando como referência dois ecossistemas (cerrado sensu stricto, que é a savana típica brasileira, e o cerradão, que é a formação florestal resultante da supressão de fogo). Em parcelas de 5 x 20 m (100 m2), com cinco repetições para cada tratamento, avaliamos densidade de plantas, área basal, altura e cobertura do dossel, riqueza e composição da comunidade vegetal e também a diversidade taxonômica e funcional das comunidades de formigas. Registramos, no total, 247 espécies de plantas, com enorme variação dos parâmetros de estrutura e riqueza entre os tipos de vegetação estudados. Plantios de árvores podem atingir a riqueza de plantas e a estrutura do Cerradão, se este for considerado o ecossistema alvo da restauração. Riqueza deste tipo de vegetação, no entanto, está muito abaixo das encontradas no cerrado sensu stricto Perdas de espécies de plantas típicas de savana, seja devido à conversão da terra ou pelo adensamento da vegetação são irreversíveis e não podem ser recuperadas por ações de restauração convencionais. Para as formigas, registramos 131 espécies, com valores de riqueza maiores nas áreas em restauração e nas referências, mas altos valores de riqueza também foram encontrados nos diferentes usos da ... / Among all savannas in the world, the Cerrado (Brazilian savanna) has the largest biodiversity and has been the most threatened by land conversion. The impacts resulting from land use changes on biodiversity have not been assessed and Cerrado restoration is a challenge still to be overcome. We investigated both - plant and ant diversity losses due to land conversion comprising production systems (sugar cane, pasture, short and long cycle forestry) and the success of different restoration techniques (passive or active, the last by planting nursery-raised seedlings of tree species in pure or mixed stands) and two reference ecosystems (cerrado sensu stricto, which is the typical Brazilian savanna, and the Cerradão, which is the forest type savanna resulting from vegetation encroachment after fire suppression). In plots of 5 x 20 m (100 m2), with five replicates per treatment, we evaluated plant density, basal area, height and canopy cover, richness and composition of the plant community, and the taxonomic and functional diversity of ant communities. A total of 247 species were recorded, with huge variation in structural and richness parameters among the vegetation types studied. Planting trees can reach the plant richness and structure of Cerradão, if this is considered the target ecosystem restoration. Richness of this vegetation type, however, are far below that existing in cerrado sensu stricto. Species losses in the typical Brazilian savanna, either due to land conversion or vegetation encroachment are irreversible and cannot be recovered by conventional restoration actions. In total 131 ant species were recorded, with greater richness values in the restoration and reference areas, but a high richness values were also found in different land uses. The ant species composition and the ant functional composition showed differences among treatments, following the complexity of vegetation ...

Orientador: Giselda Durigan / Banca: Vera Lex Engel / Banca: José Marcelo Domingues Torezan / Banca: Alexander Vicente Christianini / Banca: Elza Maria Guimarães Santos / Resumo: Entre todas as savanas do mundo, o Cerrado tem a maior biodiversidade e tem sido a mais ameaçada pela conversão de terras. Os impactos resultantes de mudanças de uso da terra sobre a biodiversidade não foram avaliados e a restauração do Cerrado é um desafio ainda a ser superado. Nós investigamos ambos - as perdas de diversidade de plantas e formigas devido à conversão da terra que compreende sistemas de produção (cana-de-açúcar, pastagem, silvicultura de ciclo curto e ciclo longo) e o sucesso de diferentes técnicas de restauração (passiva ou ativa, esta envolvendo o plantio de mudas em povoamentos puros e mistos), utilizando como referência dois ecossistemas (cerrado sensu stricto, que é a savana típica brasileira, e o "cerradão", que é a formação florestal resultante da supressão de fogo). Em parcelas de 5 x 20 m (100 m2), com cinco repetições para cada tratamento, avaliamos densidade de plantas, área basal, altura e cobertura do dossel, riqueza e composição da comunidade vegetal e também a diversidade taxonômica e funcional das comunidades de formigas. Registramos, no total, 247 espécies de plantas, com enorme variação dos parâmetros de estrutura e riqueza entre os tipos de vegetação estudados. Plantios de árvores podem atingir a riqueza de plantas e a estrutura do "Cerradão", se este for considerado o ecossistema alvo da restauração. Riqueza deste tipo de vegetação, no entanto, está muito abaixo das encontradas no cerrado sensu stricto Perdas de espécies de plantas típicas de savana, seja devido à conversão da terra ou pelo adensamento da vegetação são irreversíveis e não podem ser recuperadas por ações de restauração convencionais. Para as formigas, registramos 131 espécies, com valores de riqueza maiores nas áreas em restauração e nas referências, mas altos valores de riqueza também foram encontrados nos diferentes usos da ... / Abstract: Among all savannas in the world, the Cerrado (Brazilian savanna) has the largest biodiversity and has been the most threatened by land conversion. The impacts resulting from land use changes on biodiversity have not been assessed and Cerrado restoration is a challenge still to be overcome. We investigated both - plant and ant diversity losses due to land conversion comprising production systems (sugar cane, pasture, short and long cycle forestry) and the success of different restoration techniques (passive or active, the last by planting nursery-raised seedlings of tree species in pure or mixed stands) and two reference ecosystems (cerrado sensu stricto, which is the typical Brazilian savanna, and the "Cerradão", which is the forest type savanna resulting from vegetation encroachment after fire suppression). In plots of 5 x 20 m (100 m2), with five replicates per treatment, we evaluated plant density, basal area, height and canopy cover, richness and composition of the plant community, and the taxonomic and functional diversity of ant communities. A total of 247 species were recorded, with huge variation in structural and richness parameters among the vegetation types studied. Planting trees can reach the plant richness and structure of "Cerradão", if this is considered the target ecosystem restoration. Richness of this vegetation type, however, are far below that existing in cerrado sensu stricto. Species losses in the typical Brazilian savanna, either due to land conversion or vegetation encroachment are irreversible and cannot be recovered by conventional restoration actions. In total 131 ant species were recorded, with greater richness values in the restoration and reference areas, but a high richness values were also found in different land uses. The ant species composition and the ant functional composition showed differences among treatments, following the complexity of vegetation ... / Doutor

In the black cotton ecosystem of Laikipia, Kenya, four symbiotic ants coexist at a fine spatial scale on canopies of Senegalia drepanolobium. They exhibit different aggressive behaviours and modify their tree canopies differently. These diverse behaviours were expected to affect the associated canopy arthropod communities. At the Kenya long-term exclosure experiment (KLEE) and its immediate environs at Mpala Research Centre, Laikipia, the insect communities coexisting with each of the four ant species were characterized, and their response to different vertebrate herbivory. Other ant species inhabiting the tree canopies or the ground were surveyed too. Pitfall trapping was used in sampling terrestrial ants, while beating and mist-blowing were used in collecting arboreal insects. Different sampling methods had varying efficacies, revealing the importance of using several methods. There are at least sixteen ant species in this ecosystem, all occurring on the ground, but only ten species on the trees. Terrestrial ant communities in this ecosystem cannot be used as indicators of grazing pressure for range management. A total of 10,145 individual insects were collected from the tree canopies, comprising of 117 species from seven orders and 25 families, forming a complex community of species interacting at different levels. Symbiotic ant species had a significant effect on insect community structure and composition. Crematogaster sjostedti was associated with a community that was significantly different from the other ant species. There was no significant effect of vertebrate feeding pressure on the canopy insect community, but there was an interaction effect between ant species and treatments. Significant differences between ant species mostly occurred on treatment plots where only cows were allowed to graze. One or more of the ant species may be a keystone species in this ecosystem even though experimental manipulations failed to confirm earlier findings. It was concluded that the one-year period during which experimental manipulations were carried out was not long enough to reflect takeover effects on the insect community. The four symbiotic ant species colonizing S. drepanolobium comprises of two guilds, the hemipteran-tending ants (C. sjostedti and Crematogaster mimosae) and non-tending ants (Crematogaster nigriceps and Tetraponera penzigi). Communities associated with these guilds were found to be significantly different in all four diversity indices. The black cotton ecosystem is species-poor compared to other ecosystem such as forests. The number of insect species that colonizes S. drepanolobium and coexists with acacia-ants forms a large proportion of the invertebrate community. Therefore, this ecosystem should be conserved to safeguard this invertebrate community. This will also give scientists a chance to establish how the various insect species coexist with symbiotic ants on tree canopies.

The honeybees Apis cerana and Apis mellifera are derived from the same ancestral base about two million years ago. With speciation and evolution, they have acquired many advanced living skills in common, but have also evolved very different living strategies due to different distributions. This thesis is an intensive study of the biology of the mixed-species colonies of these species, the aims of which were to investigate their behavioural relationships and uncover the evolutionary conserved features of their behaviours subsequent to speciation. The results show that the two species can form a stable society to perform normal tasks. First, workers of both species in the mixed-colonies could form the typical retinue behaviour to hetero-species queens, thus indicating that queen pheromones could be spread to and by both species. Secondly, both species did not show significantly different ovarian activation under hetero-species queens, suggesting that the queen pheromones more likely play a role of "honest signal" rather than a "repression" substance in the honeybee colonies. Thirdly, both species could mutually decode each other‘s waggle dances, with unexpectedly low misunderstanding; revealing that the dance language in a dark environment is quite adaptive for cavity-nesting honeybees. Fourthly, workers of both species could cooperate with each other in comb construction, although the combs they built contain many irregular cells. Interestingly, A. cerana workers could be stimulated by A. mellifera workers to perform this task, thus confirming self-organization theory in the colony. Fifthly, A. mellifera workers behaved more "defectively" in thermoregulation, but perhaps because A. cerana workers are more sensitive to changes in hive temperature. Given these differences in strategy, A. mellifera workers‘ performance might in fact reduce conflicts. Lastly, when faced with threats of predatory wasps, both species engaged in aggressive defence. Although they did not learn from each other‘s responses, species-specific strategies were adopted by each of them so that the defence of the mixed-colonies is very effective. I conclude that the two species can adapt to each other‘s efforts and task allocation is reasonably organized allowing mixed-species colonies to reach stability. These results suggest that all of the social behaviours discussed here were highly conserved following speciation. This thesis could provide some clues for the study of honeybee evolution from open-nesting to the transition of cavity-nesting.

PhDENV / Department of Ecology and Resource Management / See the attached abstract below

Ropalidia marginata is characterized as a primitively eusocial wasp due to the absence of morphological differentiation between the queen and worker castes. Unlike other primitively eusocial wasps, however, the queen in this species is a docile individual, who does not use aggression to regulate worker reproduction, and does not act as the central pacemaker of her colony. However, if the queen dies or is experimentally removed, one of the workers steps up her aggression immensely within minutes, and if the queen is not replaced, she develops her ovaries, reduces aggression and takes over as the new queen of the colony. We call her the potential queen (PQ). When I started my work on R. marginata, two very intriguing questions were demanding to be answered, which had developed from work done by my immediate seniors in the lab. I decided to pursue both of these for my thesis. My work has been enriched by inputs from several collaborators and colleagues - I couldn’t have done all of it by myself. So, henceforth, I will be using the word “we”, instead of the first person singular to describe the work that has gone into this thesis. Question 1: Is there a designated successor to the queen in R. marginata? My senior Sujata P. Kardile has shown in her thesis, that in R. cyathiformis, a primitively eusocial wasp very closely related to R. marginata, the queen is always succeeded by the next most aggressive individual in the colony, and so the PQ is easily predictable in the presence of the queen. However, in R. marginata, the PQ appears to be an unspecialized individual, who cannot be predicted in the presence of the queen by using age, ovarian profile or behaviour as the yardsticks. However, the PQ becomes evident within minutes after queen removal. The swiftness with which the PQ is established led us to believe that perhaps the successor to the queen in R. marginata is known to the wasps, though we cannot identify her in the presence of the queen. We designed an experiment to check for the presence of such a ”cryptic successor” in R. marginata. Our experiments involved splitting a normal, queen-right nest into two halves separated by a wire mesh partition, so that the wasps could not move across the mesh. Earlier we had used this set-up to demonstrate that a PQ always establishes herself on the queen-less fragment of the nest. So, to test if there is a cryptic successor, we allowed a PQ to establish herself on the queen-less fragment, and then exchanged the queen and the PQ (designated as PQ1) between the two sides. There is a 50% probability that the cryptic successor, if present, would be on the queen-less side in the beginning. Then, upon exchange, she should be able to hold her position on the other side easily. On the other hand, if the cryptic successor is first on the queen-right side, then, upon exchange, she should take over as the PQ (PQ2), and PQ1 should not be able to hold her status. The cryptic successor hypothesis had two predictions: (i) the PQ1 would lose to a PQ2 in about half the cases, and (ii) there would never be a PQ3. We obtained a PQ2 in 5 out of 8 cases, and we never had a PQ3. So we could conclude that there is indeed one individual who is the designated successor to the queen in R. marginata. Since we could not identify her in the presence of the queen, we call her the cryptic successor. The cryptic successor did not receive even a single act of aggression from the PQ1, or from any other individual in the colony. Thus we conclude that she is acceptable to all the wasps in the colony. We next used the more sophisticated and rigorous method of network analysis to check if the PQ could be predicted due to some unique position she might be holding in the social network on her colony. Since this was a first study in a primitively eusocial insect using network tools, we began by characterizing the social networks of queen-right and queen-less colonies of R. marginata, and compared them with the R. cyathiformis networks to see how different the R. marginata society is from a typical primitively eusocial one. The R. marginata social networks based on dominant-subordinate interactions were low in their centrality measure as compared to the R. cyathiformis networks. However, in both the species, the queen-less networks were highly centralized, star-shaped networks with the PQs at the centre. Neither the queens, nor the PQs were key individuals in the queen-right colonies, but it is interesting to note that the removal of an insignificant node, the queen, resulted in a major change in the network architecture, converting the de-centralized queen-right network into a highly centralized one. Such centralized star-shaped networks are unique, and to our knowledge, the first ever described, in any social system. When we removed the queen from the data set (in silico removal), the resulting network was similar in centrality to the queen-right networks. We then did a comparative analysis of the positional importance of the PQs of the two species, and tried to see if we could use this as a tool to predict the PQ in the queen-right network. In R. cyathiformis, the PQs had consistently high ranks (mostly rank 2) in the network based on the degree index, while the PQs in R. marginata had random ranks in the hierarchy. However, since the PQs are known not to have unique ranks in the dominance hierarchies, we repeated the analysis using data on all interactions from the Q-PQ exchange experiments described above. Neither the cryptic successors nor the losers occupied any unique ranks in the all interactions networks. Thus the successors in R. marginata are truly cryptic, even in their social networks. Since R. marginata is known to be more evolved than typical primitively eusocial species, it is likely that the queen’s successor is identified by the wasps through some subtle cue like smell, and so we cannot identify her using the methods that are adequate for the identification of the PQ in a typical primitively eusocial species like R. cyathiformis. Question 2: How does the queen signal her presence and reproductive status to her workers or, how do the workers perceive the presence of their queen? The fact that in spite of her docility, the queen in R. marginata manages to maintain complete reproductive monopoly in her colony, gives rise to the obvious question of how she suppresses worker reproduction. The most attractive hypothesis is that she uses a pheromone like queens of highly eusocial species. My senior A. Sumana had shown that the queen pheromone, if present, is not a volatile substance. She also showed that the queen interacts at a very low rate with her workers, and so they cannot possibly perceive her by means of direct interactions. Since the PQ steps up her aggression within minutes of queen removal, we used her as a proxy to know how soon the queen’s absence is felt in the colony. We built a model to delineate the relationship between the decay time of the pheromone (td), the average age of the queen’s signal present with the PQ (ta), and the average realization time (tr); where tr = td − ta. Using Dijkstra’s algorithm, we showed that the queen could interact faster with the PQ by using relay interactions. Then using experimental data from 50 colonies, we obtained a ta of 102.9 minutes. The td was 340 minutes, and so we obtained a tr of 237.1 minutes; which meant that the PQ should not perceive the queen’s absence within 237 minutes of queen removal, if the queen pheromone is transmitted by a relay mechanism. However, from our experimental data, we had obtained a tr of 30 minutes. So we concluded that physical interactions, both direct and indirect were inadequate for the workers to perceive their queen. As we had ruled out physical interactions, we then wanted to check if it is possible that the queen applies her pheromone to the nest material, from where it is perceived by the workers when they walk or sit on the nest, or antennate the nest surface. The “rub abdomen behaviour (RA)” has been observed to be quite typical of R. marginata queens, and is not very common in the workers of the species. RA involves rubbing the ventral side of the tip of the abdomen or dragging it on the nest surface while walking. We thought that the queen might be using this behaviour to apply her pheromone on the nest material. So we characterized this behaviour using focal behaviour sampling, and found that the queen rubs her abdomen on the nest once in every 23 minutes. Since the observed tr is 30 minutes, it is quite likely that the queen uses the rub abdomen behaviour to apply her pheromone on the nest. The next step was to check for the source of the queen pheromone. We looked for glands that open near the base of the sting, and the Dufour’s gland was a good choice, as it is known to be involved in the recognition of egg-laying workers in the honeybees. We performed a bioassay in the blind using the crude extract of the Dufour’s gland (prepared in Ringer’s solution) from the queen. The Dufour’s gland extract of a randomly chosen worker and the solvent were used as controls. We found that the PQ responds to the queen’s Dufour’s gland extract by lowering her aggression to 65% of what she was showing on queen removal and before the application of the extract. However, the PQ did not change her behaviour significantly when the worker’s extract or Ringer’s solution was applied. The PQ’s reduction of aggression on application of the queen’s extract mimicked the reaction of PQ’s when the queen is re-introduced on the nest some time after removal. So we hypothesize that the Dufour’s gland is the source of the queen pheromone (signal) in R. marginata. This thesis has opened up newer questions pertaining to the power of the queen and the intricacies of the succession to power in R. marginata. For example, we need to pursue chemical analyses of the Dufour’s gland extract of R. marginata to have conclusive proof of it’s being the source of the queen pheromone. But that is perhaps suitable topic for my juniors in the lab, who can continue the tradition of beginning with questions opened up by their seniors!

The organization of complex societies requires constant information to flow between individuals. Because of their elaborated social structures and principally because of the division of labor, social insects depend on the efficacy of their information web in order to adapt the colony activity to its needs. Although many studies focused on understanding the regulation of the foraging activity, little is known about the intranidal food distribution and stock management regulation. The aim of this thesis is to quantify and describe the dynamics of the food flow and its regulation in an ant colony. A medical imagery technique, scintigraphy, was adapted to follow the propagation of radio-labeled nutrients inside the nest. This technique allowed spatiotemporal dynamics quantification of the food flow and led to the enunciation of simple yet robust regulation rules that are at work during the colony feeding process.<p>The dynamics of the harvest is regulated by the coupling of a positive and negative feedbacks. The harvest acts as both: negative and positive feedbacks. Entering food-loads trigger foragers to exploit the newly discovered food source through the well-known recruitment process. At the same time, the harvest proportionally reduces the entering food flow until the complete stop of the foraging activity when the colony reaches satiety. Surprisingly, the positive feedback (that is the recruitment) is not responsible for a faster entering food flow and is not influenced by the colony needs while the exploring activity is. The spatial dynamics of the food exchange network revealed stable patterns and fine tuning regulation of the feeding process. Spatial analysis of the food distribution showed that sucrose is heterogeneously stored among individuals and also heterogeneously consumed. We observed a regular spatial structure leading to centralization of the stocks: heavy loaded individuals being at the center of the cluster and weakly loaded individuals at its periphery.<p>The spatiotemporal quantification of the food flow allowed describing and understanding the flexibility of the colony to adapt its working force according to its nutritional requirements.<p> / Doctorat en Sciences / info:eu-repo/semantics/nonPublished