Behavioral and looking responses, and habituation of the responses, of wild-born, captive squirrel monkeys (Saimiri sciureus) to a variety of pictorial stimuli are described. The experiments were conducted using a free-viewing technique in the animals' home cage. Responses were highest to pictures of predators and prey, with lowered responses to eye-like schemata and non-associative stimuli, such as random shapes. Response habituation occurred with stimulus repetition. The data indicate that: (1) picture perception occurs; (2) certain classes of stimuli produce meaningful perceptions; (3) recognition of familiar pictorial stimuli; and (4) perception of specific stimulus features in feral squirrel monkeys.

Helping at the nest was recorded at 25% of Red-throated Bee-eater nests in Nigeria. Broods attended by pairs with helpers (multiple breeding units, or MBU's) received significantly more feedings than broods attended by unassisted pairs. Higher provisioning rates resulted in improved brood growth. In broods of three raised by pairs, the growth rate constant K of nestlings N1 and N2 was significantly higher than for N3; in broods raised by MBU's, no differences in K were found for N1, N2 and N3. Brood reduction occurred in 59% of broods of three raised by pairs, but in only 18% of MBU-raised broods. MBU's fledged 2.7 young per nest, significantly higher than the 2.0 young per nest fledged by pairs. To determine possible causes of the preponderance of males in adult populations of Red-throated Bee-eaters, adult weight loss over the nestling period, and nestling sex ratios were investigated. Males and females weighed less at the end of the nestling period than they did at the beginning, but no intersexual differences were found. The nestling sex ratio was found to be 1:1. Imbalance of the sexes was attributed to a combination of greater energetic investment to reproduction by females, and the tendency for females, rather than males, to disperse from the natal colony. Comparisons of life history parameters of the Red-throated Bee-eater with those of the Carmine, Little and European Bee-eaters, indicate that bee-eater species which are both colonial and migratory are more likely to be co-operative breeders than solitary-nesting, sedentary species. Since helpers are almost always offspring of the breeding pair they assist, helping behaviour is considered a strategy, through kin-selection, to minimise lose of individual and inclusive fitness.

The purpose of this study was to collect information on the breeding biology and life history of the Razorbill for comparison with similar studies in other species. An estimated 205,000 pairs of Razorbills breed in colonies either side of the northern Atlantic; about 70% of the birds are found in British and Irish colonies (Chapter 1). The study was conducted on the island of Skokholm, Pembrokeshire (Dyfed), South Wales between 1971 and 1973, with additional observations in 1970 and 1974. Adult Razorbills and varying numbers of chicks have been ringed on the island nearly annually since 1934 and since 1963 the chicks have been given, colour rings to denote age classes. Adults trapped in the present study were colour ringed so that they could be identified without repeated retrapping. The sexes proved impossible to distinguish on body measurements but head size provided an alternative method of sexing adults (Chapter 2). Colony attendance was recorded (Chapter 3), both during the pre-laying period and the breeding season. Breeding and non-breeding birds were separated whenever possible and diurnal and seasonal attendance at two types of colony compared. The effect of prevailing weather on attendance was evaluated. Courtship behaviour and mating frequency were recorded (Chapter 4A). Most birds returned to their natal colony to breed and remained there, usually retaining the same nest site. Birds appeared to keep the same mate so far as the annual mortality allowed (Chapter 4Bi). Laying (in May) occurred on average 10-13 days later in 1972 than in 1971 or 1973 (Chapter 4Bii). Birds advanced their laying date with age so that those breeding late in the season were mainly young birds. The little information available on the seasonal abundance of the razorbills' chief food species Ammodytes spp., suggested that the onset of breeding coincided with an increase in available food which appeared to be the main factor determining the timing of the breeding season. The development of synchronised colony attendance during the pre-laying period, was also thought to play an important part in timing breeding. Length, breadth, weight, shell thickness and pigmentation pattern were measured on samples of eggs in the 3 years. Shell thickness in 'fresh' eggs from Skokholm appeared similar to that of museum eggs collected before 1940. Shell pigmentation patterns were consistent from one year to the next in eggs laid by the same female. A seasonal decline in egg volume was recorded, apparently related to the age structure of the population. Young birds laid small eggs, but egg volume increased with age. On average 30% of all eggs laid failed to hatch and more than 70% of this loss was attributable to gulls and particularly Jackdaws Corvus monedula; the latter first bred on the island in 1965 and are increasing annually. About a quarter of the eggs lost, mainly those laid and lost early in the season, were replaced. At least some of the two egg clutches found were produced when a replacement egg was laid for a first egg which had become chilled during incubation. Incubation was on average 35 days and the nestling period averaged 17 days. About 7% of all chicks hatched failed to fledge (Chapter 4Biii). Most losses occurred in the first five days of life whilst chick weight was still related to the size of the egg from which it hatched. Behaviour at fledging (the female of the pair usually took the chick to the sea) is discussed in relation to the brevity of the nestling period and similar behaviour in other species. The frequency and composition of feeds to the chicks were examined and growth recorded by regular measurements of weight, feather length and tarsus length. Growth rates were fastest in 1973. Growth was independent of hatching date although chicks from early eggs were the heaviest at fledging.

Previous extensive ringing and the cull of Herring Gulls between 1972 and 1974 on the Isle of May by the Nature Conservancy Council facilitated an investigation of the effects of age on the breeding and recruitment of this gull species. Median laying date increased with increasing female age, four year old birds laid 11 days later than the colony median laying date, five year old gulls advanced their laying by six to nine days and laying dates progressively advanced up to the seventh year of life. Clutch size increased with the age of females. For four, five, six and seven year old birds mean clutch sizes were 1.71, 2.18, 2.43 and 2.74 eggs per female. Although clutch size decreased as the season progressed, in any time period the largest clutches were laid by the oldest birds. For all eggs laid regardless of clutch size, egg breadth and volume increased with age of the laying female. The seasonal decline in egg volume was not solely attributable to recruits laying later since four and five year old gulls laid smaller eggs than older birds regardless of the time of laying. Hatching and breeding success both increased with parental age. A large proportion of chicks raised by inexperienced breeders died within the first 10 days after hatching. This component of chick mortality was attributable to inadequate reproductive behaviour by inexperienced birds. Most four and five year old gulls nested at low densities and there was some evidence to suggest that breeding success may decrease at the highest nesting densities. The colony of Herring Gulls on the Isle of May has been increasing at about 13-14% per annum since 1950. This rate of increase is similar to other Herring Gull colonies elsewhere in Britain. Using the survival of gulls ringed as adults prior to the cull, the average annual survival rate is estimated as 0.935 - 0.100 and the survival rate in the first year of life was calculated as 0.83, 0.83 and 0.67 in three successive year classes. Prior to culling, the Isle of May was a dense colony and although some three year old gulls held territory, none bred. Some four year old gulls held territories and bred, although the proportion was low. It is estimated that the mean age of recruitment was 5.00 years and that the mean age of first breeding was 5.25 years, with 55% of birds breeding for the first time when five years old. Some individuals probably do not breed until they are seven years old. There was no evidence of differences in the age of recruitment of male and female Herring Gulls. Calculations based on the life-table of the Herring Gull and the number of marked recruits recovered during the cull lead to the conclusion that approximately 70% of the surviving young did not return to breed at their natal colony. Young reared on the Isle of May have been found nesting up to 250km away. Of these recruits which returned to nest on the Isle of May, 66% nested close to their place of birth. Males showed a greater tendency to return to their natal area than females, and this was evident both amongst those birds nesting on the island and through a higher proportion of females emigrating into other colonies. Clearing experiments suggested that recruitment was influenced by the density of breeding birds. Recruits were attracted to dense, undisturbed areas but they had much greater difficulty in establishing a territory there. A comparison of pre- and post-cull age structure on the island indicated large increases in recruitment pressure from gulls in their third, fourth and fifth years of life as a result of culling. From culled samples of known aged birds it was evident that bodyweight increased with age in both males and females. Body weight differences were correlated with increasing weights of Pectoralis major flight muscle, but not with overt visceral fat storage. Thus the breeding performance of recruits may have had both a physiological as well as a behavioural component. In mid-May recruit gulls had smaller brood patches and males had undeveloped testes in comparison to established breeders. On previously culled areas, recruits had significantly larger brood patches, a result of the greater ease of establishing territories by these birds on partially cleared breeding areas.

A detailed knowledge of the behaviour and reproduction of any exotic species of mammal plays an important role in the successful maintenance and breeding of that species in captivity. Data from the field are particularly important in evaluating results in captivity. Where complete field data are not available, problems arising in captivity can only be assessed by taking successfully breeding individuals as the norm and using them as a guideline when studying reproductive problems. This study has been designed to establish the reproductive parameters of the lesser mouse lemur, Microcebus murinus, using data from the field and successfully breeding individuals to this end. Aberrations arising in captivity were examined and an attempt was made to identify their causes. During the course of this study data have been collected on the annual variations of behavioural and physiological parameters. Information on male and female reproductive physiology has been collected and hormone assays have been conducted on urine samples from oestrous, pregnant and non-pregnant females. Data on infant birth-weight and development have also been recorded. Special attention has been paid to reproductive behaviour, particularly birth and maternal care. In addition, a study has been made of social behaviour and communication. The relationship between social behaviour, status and reproductive success has been examined. Having established the "normal" condition it was possible to examine deviations from this norm. Under certain circumstances environmental stimulation was observed to either stimulate or suppress reproduction. Suppression of reproduction is discussed in the light of physical and social environmental influences. Data on pathology have also been collected and adrenal weight and histology recorded. An attempt has been made to integrate post-mortem findings with observations on reproduction and behaviour. The normative data collected in this study are discussed in terms of their adaptive significance and the deviations from the normal condition are viewed in terms of stress. An attempt is then made to indicate the positive influence which this and similar studies could have on captive animal management.

Anopheles gambiae sensu stricto is the most efficient malaria vector in Africa. Recent advances in mapping the distribution of this vector have exploited the relationship between climatic factors and vector parameters such as growth, survival and reproduction. This work was designed to investigate the effect of temperature and humidity on the development and survival of the vector and to test the use of recently developed tools in describing its distribution. The development rate and survival of the aquatic stages of the vector were investigated at 16 constant temperatures. Adults were produced between 16 -34ºC with a peak development rate at 28ºC and peak number of adults at 22 -26ºC. Larvae survived for less than 7 days at 10º, 12º, 38 º, and 40ºC but for more than 5 weeks, at 14-18ºC without any development of adults. Laboratory models accurately predicted development times at natural breeding sites in The Gambia suggesting the applicability of the models to field situations. The survival and mortality rates of adult An. gambiae s.s. were monitored at combinations of temperatures from 0-45ºC at 5 intervals and 40%, 60%, 80% and 100% relative humidity. Survival was highest at 15-25ºC and 60-100% relative humidity. The temperature - larva development relation was used to produce a distribution map across Africa while climatic data from sites at which chromosomal forms of the insect have been found were used to map the distribution of the forms across West Africa. Climate is an important determinant of insect distribution and the use of climate and vector parameters in describing or predicting vector and disease distribution will provide a cheaper and less labour intensive tool than traditional methods.

This study, on the immature stages of the Simuliidae, was undertaken in streams draining bogs in Upper Teesdale. There were four parts to the investigation: Firstly, a qualitative survey of fourteen sites showed eleven of the thirty-five British species of Simuliids to be present. Secondly, quantitative programmes showed streams to have three types of population structure: those with larvae present mainly in winter; mainly in summer; and throughout the year. Most streams were in the latter two categories, larval numbers being lower in winter at the higher altitudes. In summer 1972, absolute population estimates were obtained from two streams by removal trapping. More larvae occurred in the lower sections of the streams and, in the streams where the winter populations were low, larval numbers were larger early in summer; this was not true of the lower altitude stream. Thirdly, cumulative frequency x size-measurement plots were made on probability paper from which the modal size-classes corresponding to the larval instars of each species could be determined. All species had seven larval instars and percentage frequency distributions for each sampling date were constructed. The detailed population biology could then be determined. Larval life was much shorter in summer at 30-40 days than in winter. Two generations were distinguished, that in winter commonly having two cohorts and that in summer often as many as four.

Pterostichus melanarius (Illiger) has been established as an important predator of slugs in the laboratory, semi-natural environments and in the field. The current method of choice for studying these predator-prey interactions is using monoclonal antibodies (MAbs) coupled with enzyme linked immunosorbent assay (ELISA). Recently, DNA-based detection methods have been suggested as a viable alternative in this area of research. DNA-based detection methods proved suitable for investigating predation by P. melanarius on economically damaging slug species. Mitochondrial DNA primers were designed for the 12S ribosomal RNA gene that were specific for Deroceras reticulatum, members of the Arion hortensis aggregate and the Arion genus. The detection limit and rate of decay of slug DNA in the beetle gut was determined using the slug-specific primers in laboratory-based feeding trials. Slug remains could be reliably detected within beetle guts for at least 24 hours following ingestion, suggesting that this technique would be suitable for detecting predation in the field. Direct comparison of immunological and DNA-based detection methods revealed that the sensitivity of MAb-ELISA is greater than amplification of prey DNA using the polymerase chain reaction (PCR). The implications of these findings are discussed. DNA-based detection methods were suitable for detecting predation by beetles on slugs in the field. The number of slug-positive beetles identified using MAb-ELISA was greater than when PCR and prey-specific primers were used. These results are discussed in the context of the availability of target and alternative prey in the field. Sequence data was also analysed to reveal the population structure and demographic history of A. hortensis aggregate in the British Isles. A high level of differentiation between populations was revealed. The rate of molecular evolution in these species is rapid with high levels of inter-specific divergence. The 12s rRNA gene proved especially useful in phylogenetic reconstruction and corroborated previous results based on morphology and enzyme electrophoresis that the congener, Arion intermedius belongs to the same subgenus (Kobelita) as the aggregate species.

This thesis addresses the problem of behaviour selection and learning in an ethological manner. This considers the agent’s perception of the environment and of its internal drives as modulatory elements that bias the way behaviours are selected. Furthermore, this happens in a framework where the agent is capable of perceiving changes in the feedback from the environment in terms of reward. A schema to learn object affordances is proposed and integrated in an actor-critic reinforcement learning algorithm. This is proposed to be the core of a motivation and reinforcement framework driving the selection of behaviour and the adaptation of behavioural patterns. Its working principle is the aforementioned capacity of perceiving changes in the environment and in the agent’s internal physiology in terms of reward, and of modifying the behavioural patterns accordingly. The aforementioned ideas on motivation, behaviour selection, learning and perception have been made explicit in an architecture integrated in an simulated robotic platform. To demonstrate the reach of its validity, extensive simulations have been performed to address the affordance learning paradigm and the reach of the adaptive power offered by the framework of the actor-critic. To this end, the effect of external and internal perception on the learning and behaviour selection processes has been measured along there different dimensions: the performance in terms of flexibility of adaptation, the physiological stability and the cycles of execution of behaviour at each contingency. In addition to this, this thesis has begun to frame the integration of behaviours of appetitive and consummatory nature in a single schema. Finally, it also contributes to disambiguate the role of dopamine as a neurotransmitter involved in the process of learning.

Experiments were performed in stream tanks and aquaris to investigate the effects of environmental complexity upon the aggressive behaviour and growth of juvenile brown trout (Salmo trutta L.) and the carrying capacity of the environment. Natural materials and various structures were used to provide environmental complexity. Aggressive interactions were found to be significantly more frequent in simple environments than in complex ones. The different types of structures used to modify the complexity of the experimental environments had different effects upon the observed levels of aggressive behavioral structures providing overhead cover were associated with the lowest levels of aggression. Increasing fish density resulted in relatively greater increase in aggression in simple environment. The wigwag display of the fish was analysed and found to be a highly ritualised, defensive manoeuvre, the efficiency of which was determined by the social ranks of the fish involved. Carrying capacity of complex environments was significantly greater then that of simple environments. There was a significant inverse correlation between aggression and resident population size in the environment types used. In the growth experiment, different relationships between individual fish's specific growth rates and their size rank position were found in populations of fish kept in simple and complex environments. The differences were probably associated with the different levels of aggression observed in the two environment types and the effect of environmental complexity upon feeding behaviour.