When foraging alone in a indoor arena, sheep adjusted their patch leaving behaviour according to their expression of patch quality. This was achieved by combining information about the average probability of reward within a patch with recent patch experience. In a supplementary experiment no significant relationship could be found between the patch leaving behaviour leaving behaviour of the sheep in this experiment and measures of personality traits for the individual animals. Under the same conditions, but paired with another sheep of known dominance status a second experiment investigated the effect of increased social complexity on patch leaving behaviour. The behaviour of individuals became less efficient as sheep of different social status diverged in behaviour. Dominant animals followed and challenged subordinate animals for patch access. Subordinate animals relinquished patch information to avoid conflict with dominants. When given a choice between predictable and variable patches offering the same mean reward in the third experiment, sheep selected that predictable patch regardless of recent patch experiences. As the time between successive patch visits increased sheep reverted to a random choice, except where recent experience on the variable patch had been negative, highlighting the persistence in memory of sheep for negative over positive or neutral experience and providing evidence for a win-stay strategy of foraging. In a final pasture experiment combining spatial, temporal and social factors, sheep foraged more efficiently when resource distribution was stable and predictable. Sheep used spatial memory to return to feeding stations within a site as a strategy to avoid competition from other flock members. This study found evidence that sheep behave flexibly and rapidly to changes in their foraging environment. Social constraints and the costs of obtaining and retaining information in a variable environment influenced efficiency in foraging behaviour.

This thesis examines in detail some of the internal and external factors which are implicated in the causation of foraging behaviour in growing pigs. Special reference is given to the interface between exploratory and feeding motivations, and the role of gastro-intestinal distension in determining the tendency to perform foraging behaviour. Four experiments were performed which found: 1. Extrinsic exploration equates to the appetitive component of the dominant motivation (e.g. extrinsic exploration for food and foraging), and therefore can be directly affected by the level of feeding motivation, whereas intrinsic exploration is not directly affected by the level of feeding motivation. 2. Chewing behaviour, such as tail-biting, is initiated through extrinsic exploratory behaviour, but then sustained by feeding motivation. In addition, feeding motivated chewing is moderately reinforced by taste, but is strongly reinforced and perpetuated by taste paired with a relevant source of feedback. 3. Growing pigs are able to respond for the repeated presentation of a feeding related conditioned stimulus in a manner which is indicative of their level of feeding motivation. This methodology allows feeding motivation to be measured in the absence of food ingestion. 4. The level of feeding motivation resulting from the ingestion of a bulky food is inversely proportional to the degree of gastro-intestinal distension. However, this effect is only temporary because feeding motivation responds primarily to the nutrient content of a meal.

Despite its potentially crucial role in improving the fit between an organism’s phenotype and the environment, relatively little is known about exactly when and how animals should use learning within their natural habitats. This thesis integrates the comparative approach with techniques traditionally developed in experimental psychology to assess how divergent habitat conditions shape learned responses in the threespine stickleback, <i>Gasterosteus aculeatus.</i> Fish collected from different habitats (ponds and rivers) were trained to learn a simple spatial task in which both landmarks and turn direction (turn left or right) reliably indicated reward location. Pond fish used both landmarks and turn direction while river fish showed a preference for using turn. In rivers, flow and turbulence may make local visual features unstable and therefore unreliable as positional cues. However, both pond and river fish fail to learn to use landmarks as goal directing cues when they are the only predictor of reward location and unstable with respect to all other sources of spatial information. A controlled rearing experiment was carried out to investigate the causal basis of cue preferences in pond and river fish. The results suggest both genetic and environmental factors may influence cue preference in the threespine stickleback. A comparison of spatial learning by sympatric species of threespine stickleback that occupy different microhabitats (pelagic and littoral zones) within the same lakes, revealed a species difference in the rate at which they learnt task. The two species may therefore be equipped with learning abilities that best suit them to either a littoral or pelagic lifestyle, even within the same macrohabitat. Together, these results suggest that learned behaviour is fine-tuned or adapted in response to local habitat conditions on a fine scale. Learning appears to operate in close conjunction with genetic and or developmental processes that enable and direct it in response to particular ecological problems.

Aspects of the behaviour of the wood ant <I>Formica aquilonia</I> were studied using semi-natural conditions in the laboratory and under natural conditions in Loch Ard Forest, Aberfoyle, Scotland. In this study the main effort was towards an understanding how the foraging population is structured and controlled. As the preferred food site of the foragers is often the first located site: maximisation of the net energetic yield of the colony as a whole may conflict with the optimal foraging as measured at the individual level. Site allegiance is a constant property of certain individuals within the colony's foraging population. A key component of the foraging system is a mechanism whereby a colony keeps a large proportion of its forager force distributed on the most profitable food sites whilst it members show strong site allegiance. Wood ants discriminate food sources without making comparisons among honeydew sources. Naive foragers show transient behaviours in their foraging repertory more often that veteran, allegiant foragers. This behaviourally flexibility of naive foragers is used to trace novel food sties likely to occur in the habitat. It is represented in this study by the small number of foragers recruited daily. By this recruiting of uncommitted foragers to the foraging population <I>F. aquilonia</I> colonies are able to overcome inherent inertia associated with site allegiance. Although significant differences may be observed at the foragers' level of activity, there are no sub populations operating at day and night time. The gradual declining activity observed when the colony was subjected to a 17.5:6.5 light:dark cycle does not begin immediately after the light out. The rhythmicity was poor in constant dark and complete arhythmicity in constant light suggesting that these rhythms are entrained by light:dark cycles. There was no evidence found to support that <I>F. aquilonia</I> foragers depend on chemical cues to orient to carbohydrate food sources: mainly honeydew provided by aphids.

Natural selection favours those individuals capable of biasing investment in male and female offspring when the reproductive value of each sex differs. One way in which parental investment can be skewed is through altering the sex ratio, i.e. the numbers of sons and daughters produced. In birds, this can potentially be achieved at the egg stage or through subsequent modification of the numbers of male and female offspring post-hatching. I investigate both sex ratio biasing mechanism in <i>Parus major</i> and <i>P. caeruleus</i> breeding on the Swedish island of Gotland. From deserted<i> P. major</i> nests where eggs had been numbered as they were laid, egg sex ratio with laying sequence increased from approximately 0.50 to 0.75. However, sample sizes were small and the effect was not statistically significant when analyses were limited to modal clutch sizes. A cross-species analysis of sex ratio variation with laying sequence suggests that females lay eggs of the larger sex first, whether that is males or females. This relationship is expected to be associated with increased levels of brood reduction. Biasing egg sex ratios in this way might increase the reproductive value of a brood if, by reducing nestling competition through mortality, the quality of remaining offspring is enhanced. There was some evidence that this may only be true for small-brooded species. Female body condition of both <i>P. major</i> and <i>P. caeruleus</i> was experimentally manipulated through heating and cooling nest boxes prior to and during egg laying. There was no evidence of alterations to egg sex ratios in relation to female body condition in either species. After hatching, higher mortality of larger male <i>P. major </i>nestlings in poor nesting conditions is expected to result in female biased sex ratios.

No description available.

A study of circadian rhythmicity of the blow fly, <I>Calliphora vicina</I>, has identified the role of locomotor activity as the 'hands' of the clock system, therefore providing a reliable gauge for the 'black box' approaches applied in the present investigation. Observations in constant conditions confirmed that the circadian rhythmicity of <I>C. vicina</I> is under the control of a self-sustained oscillatory system which is temperature-compensated and free-runs in DD with a mean period length 22.68 h. This, however, lengthened to a value greater than 24 h in continuous light below about 0.03 Wm<SUP>-2</SUP>, above this intensity, locomotor activity became arrhythmic. Phase response curves (PRCs) for 1 h light pulses and temperature steps further described the periodic changes of sensitivity of such oscillatory systems to these two main environmental Zeitgebers. The phase-shifts required for entrainment to a 24 h light/dark cycle were shown to be predictable by the light pulse PRC. The splitting of the activity rhythm into two or more components, spontaneously or induced by light stimuli, provided strong evidence that the circadian rhythm of <I>C. vicina</I> is under the control of a coupled multioscillatory system. The attempt to interpret the observation of multi-components led to the development of a heuristic model which provided support for the multioscillatory hypothesis. One of the main aims - a search for the anatomical locations of the photoreceptor(s) and the circadian pacemaker - was pursued via surgical treatments, including optic tract severance and complete removal of both optic lobes (lobectomy). Both treatments failed to destroy the activity rhythm or prevent entrainment to a light/dark cycle.

In estuaries throughout the northern hemisphere resident-freshwater forms of the threespined stickleback <i>(Gasterosteus aculeatus) </i>have undergone parallel divergence from their marine ancestors. Sticklebacks are useful for studies of reproductive isolation and speciation because divergent resident-freshwater and migratory anadromous forms exist in sympatry in the lower reaches of rivers. A great deal is already known about the evolutionary history of sticklebacks, however, little is known about reproductive isolation in wild populations. In this thesis, I investigate the nature of reproductive isolation in an anadromous-freshwater stickleback hybrid zone located in the River Tyne, Scotland and explore whether the same genes underlie variation in the same traits in different populations. 1. There was no evidence of morphotype-based assortative mating, in an experimental pond manipulation. Hybridisation between morphotypes occurred readily and this in part is due to the tendency of freshwater females to mate with large males. 2. In contrast, there is some evidence of premating isolation existing in the River Tyne wild population, and this is likely to be ecologically dependent assortative mating. Premating isolation is not strong, however, since hybrid juveniles represented 33% of the sample from sympatric sites. There was no evidence of hybridisation showing a directional bias. 3. Evidence suggests that morphological and genetic differences between anadromous and freshwater sticklebacks in the River Tyne are maintained by postmating isolation in the wild. Genetic hybrids had reduced probability of overwinter survival. There is also strong evidence for selection against intermediate morphotypes and this selection may be sex-biased against intermediate lateral plate morphs. 4. Statistical associations between loci and traits were detected and suggest both stability and flexibility in the underlying morphological divergence in sticklebacks.

The thesis describes a study of the responses to light and temperature of the common blow fly <I>Calliphora vicina</I> Robineau-Desvoidy (Diptera: Cyclorrhapha: Calliphoridae). <I>C. vicina</I> shows a long-day photoperiodic response; long photoperiods lead to the production of developing offspring, short ones cause the flies to produce diapausing offspring. It also has an endogenous circadian rhythm that can be entrained to an external <I>Zeitgeber</I>, such as a light:dark cycle. The first part of the thesis examines the responses of strains of <I>C. vicina</I> from Finland (65<SUP>o</SUP>N), England (51<SUP>o</SUP>N) and Italy (44<SUP>o</SUP>N) to identical photoperiods and temperatures. Critical daylength (the photoperiod at which the response switched from diapause to development) and diapause duration were longer in the 65<SUP>o</SUP>N strain than in the two from further south. The 65<SUP>o</SUP>N strain's photoperiodic response was relatively insensitive to temperature. In contrast, the strain from 51<SUP>o</SUP>N was temperature-sensitive; raising the temperature decreased the amount of diapause at all photoperiods. Diapause incidence and duration in all strains were influenced by environmental factors but the parameters within which these could act were set by the genetic background of the fly. The pattern of increasing diapause incidence and duration with origin at higher latitudes probably reflects a history of differential selection. Crosses between the English and Finnish strains revealed that diapause incidence was induced solely by the adult female. Both parents contributed to diapause duration. Crosses between the Fl flies and backcrosses to the parental lines suggested that hybrid females were biased in their perception of a fixed photoperiod towards their maternal line. The second part of the thesis discusses various aspects of the response to light stimuli of flies fed carotenoid deficient or depleted diets. The response of the visual system was examined directly; that of extraretinal photoreceptors was measured indirectly via behavioural observations. The locomotor activity cycle of a majority of flies reared on a carotenoid-free artificial diet was not entrainable to a light:dark (LD) cycle. Flies reared on carotenoid-deficient heart did entrain to a LD cycle but other aspects of their response suggested reduced efficiency of the extraretinal photoreceptors compared with flies reared on carotenoid-rich liver.

Mutant <i>Drosophila melanogaster</i> flies with drastically altered free-running periods of locomotor activity (<i>period</i> mutants) were used to investigate the relationship between circadian control of overt rhythm production and photoperiodic time measurement. Comparison of the free-running rhythms of wild-type, short period and long period mutants revealed that alterations in this periodicity resulted from a decrease in the active phase (short <i>period</i> mutant) or an increase in the inactive phase (long <i>period</i> mutant) when compared with wild-type. Light phase response curves were determined for light pulses of different durations which suggested that the mutants also had an altered sensitivity to light. Comparisons of the photoperiodic response of these strains revealed a remarkable similarity from which it was concluded that the <i>period</i> locus is not essential for PPTM but may be involved in a peripheral role. The light PRC data were used to model the CDL of the photoperiodic response in the wild-type, short <i>period</i> and long <i>period</i> mutants and the modelled results did not agree with the experimental ones. A feedback model for photoperiodic time measurement involving two oscillations, a pacemaker controlling overt rhythmicity and weakly coupled to a slave involved in PPTM was shown to replicate the experimental results. In conclusion, although the locomotor activity rhythm and the photoperiodic time measuring system are both governed by circadian oscillators, they are believed to be, at best, weakly coupled together so that the oscillator controlling locomotor activity may play a role in photoperiodism.