Of the seven pelican species found world wide, only the Great White Pelican (Pelecanus onocrotalus) and the Pink-backed Pelican (P. rufescens) are found in Africa. The KwaZulu-Natal province of South Africa supports only one breeding site for each of these species, and both sites represent the southern most breeding colonies for the two species in the eastern region of Africa. These nesting sites fall within the Greater St Lucia Wetland Park, and are afforded a degree of protection, but the same is not true for their foraging and dispersal ranges, and this is a cause for concern. Considerable amounts of data have been collected relating to the status, distribution and breeding efforts of these two species in north eastern KwaZulu-Natal. These data have been collected in a variety of ways by a large number of people. Part of this work represents an attempt to collate and summarise these data to produce an estimation of the status, distribution and breeding success of these species, and to evaluate any trends in their demography. For this south eastern region of Africa I estimated the population for the Great White Pelican to range between 6000 and 9000 individuals, and the Pink-backed Pelican to range between 600 and 900 individuals. Pelicans are highly mobile birds, and this allows them to move considerable distances when they forage, disperse or migrate. They are also long-lived birds with few natural predators. The two pelican species in south eastern Africa have been poorly studied and little is known about their movements, population dynamics and causes of mortality. Habitat change poses a potential threat to pelicans in north eastern KwaZulu-Natal, and habitat loss could drive these species out this region to areas north of South Africa. Much of this north eastern region of KwaZulu-Natal is under threat, mainly through the actions of man. Many areas are naturally unsuitable for pelican foraging, while others are vital to the survival of both species. This study includes an attempt to assess the movements of these two species in south eastern Africa, and to assess the relative importance and condition of the potential pelican habitat in the north eastern KwaZulu-Natal region, focusing particularly on Lake St Lucia and the Pongolo River floodplain. All this is necessary to produce a baseline from which long term predictions of potential pelican species survival can be made. In the absence of documented life tables and environmental variability data, a range of parameters was modelled to generate population viability analyses to simulate possible scenarios. These extinction models show the outcomes of both the deterministic and the stochastic processes. An attempt was also made to identify the factors that impact most severely on the persistence of these two species. The models were most sensitive to variation in survivorship in the first year of life and to the frequency of catastrophes. Changes in these parameters had the greatest effect on extinction risk. In January 2004 Lake St Lucia was reduced to a fraction of its normal capacity as a result of a severe drought in this region of KwaZulu-Natal. After rains in the area the lake level rose and then fluctuated considerably over the next 24 months. During this time the mouth of the estuary into the sea was closed. Great White Pelican numbers and lake levels were monitored throughout this period. This part of the study relates the changes in population numbers to the lake conditions, and highlights the importance of the lake to this avian species. It uses lake levels as a proxy for the conditions of wetlands in the Lake St Lucia region. It also addresses the implications of these relationships to the management strategy of the lake and the conservation of some avifauna. To identify conservation concerns for the Great White and Pink-backed Pelicans it was necessary to generate these baseline estimations. Although much of this information is uncertain for these two species, an attempt has been made here to predict the persistence of these species in north eastern KwaZulu-Natal and to highlight the conservation issues related to their future. / Thesis (M.Sc.)-University of KwaZulu-Natal, Pietermaritzburg, 2006

Thesis (MA (Afrikaans and Dutch))—University of Stellenbosch, 2007. / This study investigates the way in which the representation of history and of the "other" coheres with the construction of the identity of the characters in Eilande (2002) by Dan Sleigh and Pelican Bay (2002) by Nelleke Noordervliet. Initially the research contextualises the authors and their novels by means of a short biographical introduction on the two authors and their different oeuvre’s, as well as an outline of the reception history of the two chosen postcolonial novels. The theoretical background which follows, gives a review on postcolonial literature in general and, more specific, in the Netherlands and South Africa respectively. Different perspectives on representation of the history and the representation of the "other" are provided, with specific attention to the relationship between hybridity, syncretism and identity. These theoretical insights form the basis for the analysis of the two novels, where the relationships and differences between these two postcolonial texts are highlighted. The conclusion is that Eilande and Pelican Bay could be seen as, firstly postcolonial, as both novels provide criticism on the colonial order in different ways and, secondly, were published during a period which followed the colonial era. Hybridity is an important term in this study, with reference to the forming of identity of some of the prominent characters. The conclusion is that the term dynamic cultural hybridity could be used in discussions about postcolonial mixed identities. Although both novels use techniques to give a voice to voiceless marginal subjects, shortcomings in their portrayal as fully-fledged characters do, however, exist. Both novels also have as an ultimate objective to prove that it is impossible to know history and the "other" and that no one truth exists. The motif of islands which appears frequently in both novels is related to aspects like identity, exile and alienation. In conclusion it is not only apparent how strongly the different characters in the novels are formed by their socio-historical contexts, but also from the approach of the authors in their contemporary criticism on oppressing systems like the VOC and/or slavery.

La cognition physique correspond à l’ensemble des connaissances que nous possédons sur les objets inanimés qui nous entourent, et à leurs relations avec l’environnement. Selon l’hypothèse de l’intelligence sociale générale, la cognition physique se serait développée tel un continuum (à l’instar de tous les autres domaines cognitifs), chez les espèces vivant au sein de groupes sociaux complexes. A l’inverse, l’hypothèse de l’intelligence sociale adaptée suppose que le fait d’interagir avec des congénères a permis de développer des capacités cognitives supérieures, mais spécifiques, du domaine social. Bien que les recherches relatives à l’évolution de la cognition physique se soient d’abord focalisées sur les primates, nous savons aujourd’hui que certains oiseaux sont capables d’interagir de manière complexe avec leur monde physique, en utilisant et en fabriquant des outils (p. ex., voir Article 1). Néanmoins, de nombreuses familles aviaires restent à ce jour non étudiées, laissant ce débat en suspens. L’objectif général de ce travail de recherche est de contribuer à une meilleure compréhension des facteurs responsables de l’évolution de la cognition, grâce à l’apport théorique de l’hypothèse de l’intelligence sociale suivant son aspect général et adapté. Quatre espèces, jusqu’alors inconnues sur le plan cognitif, ont été sélectionnées suivant leur degré de socialité mais également la spécificité de leur écologie. Ce dernier facteur, trop souvent ignoré en biologie du comportement, est crucial pour apprécier le comportement dans son ensemble. Nous avons ainsi estimé la faculté des sujets, à savoir des pélicans blancs Pelecanus onocrotalus (sociaux) et des euplectes vorabés Euplectes afer afer (grégaires), à reproduire le comportement d’un congénère (domaine social) grâce à un test d’apprentissage social (Articles 2 part I & 3). Nous avons ensuite évalué leur capacité à se servir d’un objet pour obtenir une récompense alimentaire hors de portée (domaine physique), grâce à un test d’utilisation d’outils (Articles 2 part II & 4). Bien que les pélicans fussent capables de résoudre rapidement la tâche d’apprentissage social, ils ne réussirent pas à utiliser spontanément des outils (cf. discussion Article 2 part II). A l’instar du pélican, l’euplecte imita le comportement d’un congénère. Néanmoins, il échoua à utiliser des outils dans le contexte du fourragement, malgré le fait que cet oiseau utilise et fabrique des outils de manière complexe pour construire son nid.A défaut d’avoir pu étudier l’apprentissage social et l’utilisation d’outils chez deux autres espèces sociales, les calaos terrestres Bucorvus et les toucans Ramphastidae, cette recherche consistait également à administrer un paradigme permettant d’apprécier le domaine physique : le test de la ficelle (Articles 5 & 6, respectivement). La tâche impliquait de tirer sur une ficelle afin d’obtenir une récompense alimentaire accrochée à son extrémité. Les calaos terrestres échouèrent à tirer sur la ficelle dans la configuration verticale, mais réussirent rapidement la tâche dans diverses conditions de la configuration horizontale. Chez les toucans, cependant, un seul sujet réussit le test dans sa configuration verticale. Les résultats obtenus nous permettent de réfuter l’hypothèse de l’intelligence sociale dans son aspect général. En effet, aucun lien ne semble se dessiner entre le domaine social et physique chez les quatre espèces aviaires étudiées. Bien que l’hypothèse de l’intelligence sociale soit soutenue depuis plus de quatre décennies, une théorie unitaire est requise. Dans ce cadre, un nouveau modèle d’évolution cognitive, permettant d’évaluer l’importance de l’intelligence générale chez une espèce donnée, pourrait s’avérer particulièrement prometteur. / Physical cognition is defined as the knowledge that we possess about the inanimate objects surrounding us, and their relation with the environment. According to the general social intelligence hypothesis, physical cognition would have developed as a continuum (like all the other cognitive domains), in species living in complex social groups. By contrast, the adapted social intelligence hypothesis assumes that interacting with conspecifics has allowed development of superior, but specific, cognitive capacities related to the social realm. Although research on the evolution of cognition first focused on primates, we now know that some avian species are capable of interacting with their physical world in a complex way by using and manufacturing tools (e.g., Article 1). However, to date, various bird families are still unstudied, leaving open this debate. The general goal of this work is to contribute to a better understanding of the factors acting on the evolution of cognition, thanks to the theoretical input of the social intelligence hypothesis according to its general and adapted aspect. Four species that had never been studied in cognitive studies before were selected according to their degree of sociality but also according to the specificity of their ecology. This latter factor has been largely ignored in behavioural biology, although it is crucial for a more holistic comprehension of the behaviour. This work aimed to assess the ability of two avian species, great white pelicans Pelecanus onocrotalus (social birds) and yellow-crowned bishops Euplectes afer afer (gregarious birds), to imitate the behaviour of a trained conspecific (social domain) with a social learning task (Articles 2 part I & 3, respectively). Subsequently, we have studied heir ability to use an object in order to get an out-of-reach food reward (physical domain) with a tool use task (Articles 2 part II & 4). Although pelicans were capable of rapidly solving the social learning task, they did not succeed in using tools spontaneously (cf. discussion Article 2 part II). The bishops were able to imitate the behaviour of a conspecific, however they were not capable of using tools in the foraging context, although these birds are well known to use and manufacture tools in quite a complex way in order to build their nests. We have also administrated to two other avian social species, ground-hornbills Bucorvus and toucans Ramphastidae, an experimental paradigm to assess cognition in the physical domain: the string-pulling test (Articles 5 & 6, respectively). The task involved pulling on a string in order to obtain a food reward attached to its extremity. Ground-hornbills failed to pull on the string in the vertical configuration, but rapidly solved the task in various conditions within the horizontal configuration. In toucans, however, only one subject succeeded in the vertical configuration. These results allow us to refute the social intelligence hypothesis in its general aspect. Indeed, no link seems to be drawn between the social and the physical domains in the four species studied. Although the social intelligence hypothesis is supported since decades, a unitary theory is required. A new model of cognitive evolution, that allows assessing the importance of general intelligence in species, may be particularly promising.