Nature is rich with examples of the collective motion of animal groups, such as flocks of birds or shoals of fish. The mechanisms of self-organization resulting in these spectacular phenomena have received wide attention. Individual-based models are a popular and promising approach to investigate and explain features of animal collective motion. The first part of this thesis gradually develops a novel modelling framework for the collective motion of animals and justifies it by comparison to empirical findings. Key aspects of the model are stochastic asynchronous updates and sensory zone sampling of individuals. Higher updating frequencies are related to increased levels of perceived threat and reduced stochastic effects leading to synchronisation in moving groups. Sensory zone sampling, biased according to the distance between individuals, provides a possible mechanism for the empirical finding that animals on average only interact with a fixed number of nearest neighbours. Many group-living animals show social preferences for relatives, familiar conspecifics or individuals of similar attributes, such as size. How such preferences could affect the collective motion of animal groups has been rather unexplored. The second part of this thesis reviews previous theoretical work that combines the concepts of social networks and collective motion. Although many of the models reviewed have not been explored with ecology in mind, they present a current context in which a biologically relevant theory can be developed. The modelling framework from the first part is extended to include social connections as preferential reactions between individuals. This is used to show that the structure of social networks could influence the cohesion of groups, the spatial position of individuals within groups and the hierarchical dynamics within such groups. This thesis aims to contribute possible mechanisms, testable hypotheses and an informed starting point for future research on how social networks could affect animal group movement.

The triommatidium is a group of three ocular facets, mounted on a tuberole and appended to the postero-lateral part of the compound eye in most aphids. It is derived from compound eye facets, differentiates precociously in the embryo and appears to act as a functionally independent receptor in the adult. Reorientation and probing in apterous, female A, fabae was examined on the fla1 and on wax ridges. To reach a black pillar on the flat, nymphs and adults less than 90 mins. old followed sinuous paths, adults over 12 hrs. old followed simple curved paths and adults between 3 and 8 hrs. old followed paths which partially orbited the pillar, without ever reaching it. Extirpation of the triommatidium prevented the paths previously typical of nymphs and adults under I2 hrs. old and it was assumed that stimulation of the triommatidium was involved in the expression of these behaviour patterns. Probing was apparently attributable to a critical angle subtended at the compound eye by an image dead ahead and the same response was seen in aphids walking , along ridges. Aphids on ridges made abrupt turns and climbed down from the ridge when the angular acceleration of an image across the compound eye reached a critical value. The movement of an image into or out of the triommatidial field of view apparent17 elicited probing. It was inferred that the compound eyes and the triommatidia comprise functionally separate visual systems which mediate different responses in walking aphids when simplified visual situations are presented under controlled conditions. A "compromise" response apparently occurs in adults between 3 and 8 hrs. old which leads to the avoidance of visually salient features in the aphid's vicinity and may contribute towards dispersal behaviour in young adult apterae.

The evolution of species interactions is a fascinating subject, and one of vital importance if we are to understand how biological communities change over time. This thesis considers the interaction between aphids (Homoptera) and ants (Formicidae). Ants tend aphids for sugary honeydew and in return provide a variety of protective services. A literature review in Chapter 1 introduces the subject and provides background information. Chapter 2 considers ant- aphid interactions in a community setting. Specifically, I consider the fitness effects of the ant- aphid interaction on host plants. Net benefits or costs to plants depend on the densities of ants and aphids; these densities may themselves change depending on context dependent factors. Chapters 3 and 4 consider how semiochemicals can allow species to respectively maintain or avoid synchrony in space and time with mutualists or antagonists. Chapter 3 shows ladybirds avoid prey patches guarded by ants by reducing oviposition in response to ant semiochemicals. Chapter 4 shows that aphid walking dispersal can be limited by ant semiochemicals. This may be adaptive for aphids to remain in areas of enemy- free space. Alternatively, if levels of kin competition are high limited dispersal could be costly to aphids. In Chapter 5 I consider interactions between invasive and native ants. Ecological dominance in ants may be mediated by the ability to monopolise honeydew- producing resources. Chapter 6 explores ants’ decisions whether to tend or prey upon aphids. Predation of aphids depends on colony demand (e.g. through cues from the presence of larvae) as well as the quality or quantity of supply (e.g. increased predation of unproductive aphids). Finally, Chapter 7 deals with macroevolutionary patterns in the interaction between ants and aphids. Specifically, I identify ecological traits that characterise aphid- tending ants. A final discussion chapter summarises how ant-aphid interactions fit into existing mutualism theory.

Mosquito transmitted viruses (arboviruses) cause significant burden in much of the developing world. Little is known about mosquito responses to viral infection, and how these responses could be utilised to prevent spread of viral disease. Anopheles gambiae, the principal vector of human malaria, unusually transmits virtually no arboviruses, with one known exception - O’nyony-nyong Virus (ONNV). In this thesis the interactions between ONNV and the A. gambiae immune system were studied. Initially ONNV infection in A. gambiae mosquitoes and an A. gambiae derived cell line were characterised. The in vivo transcriptional responses of A. gambiae to viral infection were profiled using full genome microarrays, describing the global response to ONNV infection. This thesis demonstrates that the A. gambiae immune system does respond to viral infection, with genes covering all aspects of immunity being differentially regulated, from pathogen recognition to modulation of immune signalling, complement-mediated lysis/opsonisation and immune effector mechanisms. Furthermore, this study identified four immune genes (a galectin, an MD2-like receptor and two lysozymes) regulated by ONNV infection that are capable of limiting virus during infection. These genes have novel roles in anti-viral immunity, and suggest previously uncharacterised mechanisms for targeting viral infection. Additionally, it is shown that A. gambiae uses a combination of core conserved anti-viral mechanisms, including RNAi, but does not utilise some signalling pathways reported to be anti-viral in other insects. This indicates that species specific mechanisms target viral infection. Finally this study demonstrates that foreign RNA acts as a Pathogen Associated Molecular Pattern (PAMP) in A. gambiae derived cells, and triggers transcriptional responses that dramatically reduce viral infection. In conclusion the data presented in this thesis demonstrate that A. gambiae responds to and is capable of limiting viral infection through conserved and novel immune mechanisms triggered by recognition of viral infection and foreign RNA.

The host-plant acceptance behaviour of aphids was examined, in two aphid species, the bird cherry-oat aphid, Rhopalosiphum padi and the pea aphid, Acyrthosiphum pisum, to elucidate the recognition factors used to assess plant suitability for parthenogenetic reproduction. In addition, host-plant chemistry was investigated to study possible cues that initiated reproduction. The probing and parturition behaviour of R. padi on barley, Hordeum vulgare, or bird cherry leaves, Prunus padus, were monitored by electrical penetration graph (EPG) coupled with a simultaneous video recording. The autumn winged gynoparae and the summer winged virginoparae initiated reproduction on their host plants, P. padus and H. vulgare, respectively, before phloem contact occurred and these findings suggest that phloem contact is not necessary for host-acceptance decisions. Host acceptance behaviour of A. pisum, on susceptible and phloem-based resistant line of Medicago truncatula was monitored and found to be similar on both plant lines. The phloem-based resistance mechanism did not affect parturition behaviour indicating that reproduction is initiated prior to phloem contact. The results also imply that sign chemicals used as host recognition cues located in a peripheral tissue rather than in the phloem of host plants. Bioassays were employed to characterise the sign chemicals within host-plants of R. padi. It was found that virginoparae reproduced similarly on a neutral medium such as water or holidic diet as well as host-plant aqueous extract, which suggests that particular stimulants may be not involved in host-plant recognition by this generalist morph. On the other hand, aqueous extracts of bird cherry leaves specifically stimulated parturition in autumn gynoparae. Bioassay-guided chemical fractionation showed that at least five secondary metabolites appear to be involved in host recognition by gynoparae.

No description available.

No description available.

This thesis aimed to detennine the interactions between social structure, genetic relationships and individual behaviour in a passerine bird, the vinous-throated parrotbill Paradoxornis webbianus. During the breeding season, nest dispersion was not random in tenns of distribution pattern or genetic relationship; more related males tended to breed closely together, especially in tre second laying peak. This relatedness was positively related to subsequent juvenile recruitment (Chapter 2). Local breeding density also influenced parental provisioning behaviour. When they had fewer neighbours, parents tended to coordinate their provisioning and the provisioning rate increased as local breeding density increased. Furthennore, parents breeding in the high local density appeared to meet more efficiently an experimental increase in chick demand (Chapter 3). However, parental provisioning behaviour was not affected by the extrapair paternity (EPP). Furthennore, the pattern of EPP (8% of offspring in 26% of broods) was not related to either local breeding density, breeding synchrony or male characteristic. However, most EPP (95%) occurred in the first laying peak and extrapair males were not always neighbouring males. In addition, there was no difference between within-pair and extrapair chicks in body condition and recruitment rate to their first winter flock (Chapter 4). Juveniles showed moderate rate of dispersal; about 40% of juveniles moved from natal flocks to neighbouring winter flocks and sibling coalitions including both sexes seemed to be the rule in these movements (Chapter 5). Once juveniles settled in their first winter flocks, they had a strong fidelity to the winter flocks throughout their life (Chapter 6). Overall, these demographic patterns generated significant genetic differentiation and a pattern of isolation by distance among winter flocks at a fine spatial scale and kin association within a winter flock. Despite these demographic and genetic circumstances, however, inbred mating was relatively infrequent (Chapter 6).

The shape, size and number of sperm males produce and the characteristics of the reproductive tract through which they must travel have been shaped by a multitude of evolutionary forces. The selective pressures acting on sperm and the female tract have produced a wide variety of forms, some of which are unusual to the point that they appear maladaptive. This thesis uses a range of behavioural, physiological, morphometric and theoretical techniques to investigate the evolutionary causes and consequences of non-fertile sperm and spermicide in the Drosophila obscura species group. The primary aim of this thesis is to determine the adaptive significance of the non-fertile 'parasperm' found throughout this taxon. Parasperm have been proposed to perform a number of functions including depressing female remating behaviour and transferring nutriment, but support has not yet been found for any existing hypothesis. Similarly, many roles for spermicide have been proposed, but its adaptive significance and prevalence remain unknown. Chapter 1 discusses concepts and empirical results relevant to the background of the thesis. I first introduce a new hypothesis proposing that selection on males to mitigate the effects of spermicide can lead to the evolution of parasperm (chapter ' 2). This hypothesis is mathematically investigated and its predictions determined. I also review the previously undeveloped hypothesis that parasperm evolved in response to cryptic female choice.

No description available.