1) This thesis concerns the ecology of urban magpies Pica pica L, as investigated in an area of South Manchester. 2) Using nest census results from an area of 47 km 2 , the density of magpies was estimated as 5.9 - 6.8 pairs km ; the principal objective of the study was to explain this high density. 3) The study area was surveyed five times between 1976 and 1979; data collected included information on the preferred nest sites, nesting density, and number of magpies present. In four of the five surveys, the birds observed accounted for 69% of the magpies estimated to be present from the number of occupied nests. Nesting density was s~own to be significantly correlated with tree diversity and area of grassland. There may have been a slight increase in the number of breeding magpies between 1977 and 1978. Average fledging success in the area was estimated, from a survey in June 1977, to be 1.5 young per pair. 4) In studying the breeding ecology between 1977 and 1979, 159 nests were investigated. The average first egg date was l7th.April, and the average clutch size was 5.5 eggs. An incubation period of 24 days was calculated from the laying of the first egg, or 18 days from the laying of the last egg, to the hatching of the first pullus. Overall 55% of the pairs investigated (if replacement clutches are included), successfully raised at least one pullus to fledging. Asynchronous hatching occurs. Of the nests in which eggs were laid, 36% failed to hatch any young. The significance of partial and total failures during the egg stage are discussed. It was necessary to calculate a range for hatching success (43% - 65%) because it was not possible to determine whether the drop in reproductive potential which occured at this time, was due to loss of eggs or nestlings. Hatching success and total failures are considered in terms of seasonal variation; differences between original and replacement clutches are also investigated. Of the nests which successfully hatched young, the average brood size at fledging was 1.9 young for the original attempts, and 1.5 in replacement attempts; 34% failed to fledge any pulli. Nestling starvation was strongly indicated as the major Cause of failure, although the effects of predation could not be ruled out. 5) The high density of magp!es in Manchester corresponds to an abnormally low breeding success. If national mortality figures apply, the population should be decreasing by 16% a year. The census figures indicated that this was not the case, therefore immigration, or a lower adult mortality occurs in Manchester. 6) Increases in magpie weight, tarsus length, and ~enth primary length during the nestling period have been described using logistic growth curves. 7) The significance of selective starvation in regulating brood size is considered for a small sample of 6 nests, which were studied in detail. This study also suggests that magpies exhibit synchronous fledging. 8) Body component growth, and overall body composition in nestling magpies may be interpreted in terms of an adaptive growth hypothesis; items which have the current highest functional priOrity grow the most rapidly. It is also shown that the growth strategy of nestling magpies exhibits all the features proposed by O'Connor (1978a) as exemplifying a typical brood reduction strategist. 9) The diets of adult and nestling magpies have been interpreted from analyses of a Variety of samples. Seasonal Variation in the availability of the invertebrate fauna is represented by the results of pit falling at two sites. The nestling and adult diets are extremely varied, but are largely composed of material that is typical of grassland habitat. 10) The main points of the investigation are brought together in the conclusion to provide some indication of why the magpie is so abundant in Manchester.

No description available.

Crows (Corvus cornix and C. corone) nest on power-poles throughout northern Scotland causing interruptions to electricity that cost the electricity provider more than £250,000 annually. I aimed to establish what factors influence the nest site selection of crows in the islands of Orkney to help determine ways to reduce pole nesting by these birds. Crows preferred nest sites in trees that were coniferous, tall, mature, densely grouped, and far from occupied houses. However, when such trees were scarce, the number of occupied houses was high, and there were no cliffs within 1km, crows were more likely to nest on power-poles. I found that the fitting of a Firefly diverter at sites where nests were removed was ineffective at deterring rebuilding, but rebuilding was less likely to occur the later in the season that nests were removed. Nests in the middle phase of construction were the most likely to be rebuilt. Making an appropriate decision as to when to remove a crow nest, therefore, would seem more effective for deterring nest rebuilding than is the fitting of Firefly diverters. Nest removal also reduced the level of pole nesting both within years and between years. Furthermore, planned interruptions led to fewer customers losing power and to a twelve-fold reduction in the number of minutes that customers were without power, relative to unplanned power-cuts caused by nests. I also attempted to determine the efficacy of providing alternative nesting platforms and of insulating the live wires on transformer boxes, but I collected too few data to confirm if either of these might reduce the number of power-cuts caused by crows' nests. As both nest platforms and insulation remain in place, however, data collected from these sites in the future may allow confirmation (or not) as to their value.

It is often suggested that species differences in neophobia are related to differences in feeding or habitat specialisation. Generalist species, which have more to gain from exploring novel resources, tend to be less neophobic than specialists. However, some successful generalists including ravens, brown rats and coyotes also demonstrate high levels of neophobia. I explored this paradox using common magpies, a widespread generalist opportunist that displays behaviour indicative of high neophobia. Using a combination of field and short-term captive studies, I investigated whether novelty reactions were a fixed trait or varied according to object features and context as well as for different categories of novelty (i.e. objects, food, location). I found that novelty reactions in magpies were not influenced by object features such as colour, shape or size but varied greatly depending on environmental context and novelty category. Birds did not show avoidance of novel objects presented in novel environments but were extremely wary of similar novel objects presented in familiar environments, suggesting that violation of expectations may be more important than absolute novelty. Magpies could overcome the neophobia through repeated exploration of the objects over longer periods of time, but it affected their foraging behaviour. To avoid interactions with novel objects, wild-living magpies successfully employed an innovative technique that involves observing and pilfering from caching squirrels. Less aversion was shown towards novel food than to novel objects, while familiar objects and food encountered in novel locations were generally accepted. In total, this thesis suggests that neophobia is a complex and dynamic phenomenon in generalist foragers which may set protective limitations on the level of exploration. It can be overcome through learning and the development of alternative behavioural tactics such as kleptoparasitism.

Most of the fieldwork for this study was conducted on the island of Islay, in the Inner Hebrides, the stronghold of the chough, Pyrrhocorax pyrrhocorax L., in Scotland. The aims of this study were to provide baseline data on the phenology of potential invertebrate foods of the chough, and to provide a greater understanding of the factors affecting these invertebrate populations. The literature concerning (a) the chough in Britain and its feeding ecology, (b) the invertebrate fauna of pasture, (c) the invertebrates associated with cow dung, (d) ivermectin and its effect on the invertebrate fauna associated with cattle dung, and (e) the multivariate analysis methods used in this study, is reviewed. An area of heather moorland and four pastures were selected on Islay. Invertebrates were collected from these sites between January 1988 and November 1989 using pitfall traps, and by sampling soil and cow pats. The data obtained was analysed using two multivariate analysis methods -Two-Way-Indicator-Species- Analysis (TWINSPAN) and Detrended Correspondence Analysis (DECORANA). Information on 62 surface-active taxa was obtained from pitfall trapping. Although seasonal taxa assemblages were recognized, the distribution of the invertebrate communities was primarily related to soil moisture content. Grazing intensity and seasonality were also important factors determining the composition of the invertebrate fauna at each site. The taxa active during the summer and winter at the two sand grassland sites, did not appear to differ as markedly as at the other sites sampled. Figures showing the seasonal activity of some of the frequently occurring taxa at each site considered potential chough prey items are provided. Surface-active potential chough food items were present, at all the sites investigated, throughout the year. Soil-sampling provided information on 34 taxa. As with the surface-active fauna, the primary factor influencing the soil fauna was soil moisture content. The time of year was also an important factor governing the soil fauna composition, with the majority of taxa occurring in low numbers during the summer months at all the sites sampled. Figures indicating the seasonal occurrences of some of the taxa considered potential chough prey items at each site are provided. Soil did not appear to be a good source of potential prey items for the chough during the summer months, although, as a result of seasonal increases in size, certain taxa, e.g. Tipulidae larvae, may have been more 'worthwhile' prey items at this time of year than at any other. Information on 54 taxa was obtained from sampling cow dung. Seasonality and age of the dung were very important in determining the composition of the dung fauna. The seasonal variations in the fauna associated with the cow pats are described. Potential chough prey items were associated with cow dung, in any stage of decay, throughout most of the year. Only during the period from October/November to January did there appear to be a lack of suitably sized potential prey items in the dung. The 'summer' months, when fresh dung contained large numbers of beetle adults and developing fly larvae, and late autumn, when pats deposited during the summer months are old enough for the large numbers of Aphodius spp. larvae present to have attained a reasonable size, were considered to be the times at which cow dung presented the best feeding opportunities for the chough. Fifty taxa were identified in samples of chough faeces. Multivariate analysis of these data indicated that the seasonal availability of prey items was the most important factor influencing chough diet throughout the year. Soil-dwelling Tipulidae (January to July) and Bibionidae (January to April) larvae, dung-associated insects (during the spring, and late summer and autumn), and surface-active insects (during the summer) were important invertebrate components of the diet. Cereal grains were extremely important supplementary food items during the early winter months, when invertebrate availability was low. An experiment was conducted at the College to investigate the effects on the dung fauna of spiking cow dung with 2.0, 1.0, or 0.5 mg/kg dung of ivermectin. Pats were placed on pasture between May and September 1988. The pats were lifted, and the soil beneath sampled, after 15 to 90 days exposure. A total of 65 taxa were identified. These data were analysed using TWINSPAN, DECORANA and Canonical Correspondence Analysis (CANOCO). The major factors determining the invertebrate fauna of the pats were length of exposure, exposure date, and ivermectin presence/absence. Ivermectin markedly affected the fauna associated with the pats. Pats exposed in June and August degraded faster than those exposed in May or September. In June, the ivermectin-treated pats degraded significantly slower than the control pats. An attempt to extract ivermectin from cow dung for analysis by high-performance liquid chromatography is described. This proved unsuccessful and the reasons for this failure, and possible improvements, are discussed. The main conclusions of this study are: (1) that Tipulidae larvae are extremely important components of the chough's diet on Islay, and that the climatic conditions of the island favour these insects; (2) livestock farming on Islay, especially the out-wintering of cattle, provides essential feeding opportunities for the chough, as, (a) gazing animals produce the short sward preferred by the chough as a feeding habitat, (b) large numbers of insects are associated with the dung of these animals, and (c) supplementary feed provided for the cattle in winter also provides an essential alternative food source for the chough at a critical time; (3) the chough's preference on Islay for feeding in sandy, coastal pasture is due to the fact that these sites, (a) contain a variety of suitable invertebrate prey items throughout most of the year, (b) are normally intensively grazed and so contain large amounts of dung with its associated fauna, and (c) are used for out-wintering cattle and therefore cereal grains can be found there; (4) treating cattle with ivermectin could have an adverse effect on the chough as it reduces the number and variety of invertebrates associated with the dung, an important food source for the birds, especially in spring and autumn.