Shifting climate patterns are a cause for concern for natural ecosystems globally. Of particular concern is the effect of climate change on fire-prone, Mediterranean-type shrublands globally because of the heightened sensitivity of post-fire vegetation to environmental conditions. In this thesis, I focused on investigating the relationships between rainfall seasonality patterns and post-fire vegetation processes in neighbouring Fynbos and Renosterveld shrubland communities within the mega-diverse Cape Floristic Region of South Africa. I investigated vegetation sensitivity to moisture availability at multiple levels of detail including 1) productivity and community structure, 2) growth form responses and 3) physiological performance over three years. Post-fire rainfall patterns were manipulated by artificially increasing summer rainfall and reducing winter rainfall over permanent, field sites, thus reducing annual seasonality and creating soil moisture contrasts between control and treatment plots over warm and cool seasonal periods. At all levels of investigation, postfire vegetation processes at the Fynbos site were relatively insensitive to variations in moisture availability relative to the Renosterveld site where vegetation processes and community structure were strongly affected. Nutrient limitation and lower soil tension in coarse, sandstone-derived soils of the Fynbos site could strongly limit the influence of soil moisture patterns on post-fire physiology leading to stable growth, community structure and productivity under a variety of moisture regimes. Soil moisture patterns during the first summer had significant and long-term implications for community structure and productivity patterns in the Renosterveld site, highlighting the sensitivity of vegetation patterns to early post-fire processes. Overall this study demonstrates that post-fire rainfall patterns can have strong effects on vegetation recovery processes but that structurally similar shrublands, which are specialised to differing soil types, could show marked differences in their response to climate change due to the mediation of climate responses by soils.

South Africa is known for its highly endemic and remarkably species-rich fauna and flora. The Greater Cape Floristic Region, consisting of the Cape Floristic Region and the Succulent Karoo Region, is home to South Africa's second most diverse plant family, the Aizoaceae. While the Aizoaceae are relatively well studied, there has been little work done on their associated insect fauna. Preliminary observations in recent years revealed a diverse community of gall midges on these plants. This study aimed to investigate the gall midges associated with two speciose genera of Aizoaceae and it is the first of its kind in South Africa. I report five new species of gall midges, three belonging to Asphondylia Loew and two to Lasioptera Meigen. These species are described from several species within the Aizoaceae genera Drosanthemum Schwantes and Malephora N.E.Br., and additional host records were recorded from Lampranthus N.E.Br., Carpobrotus N.E.br., Acrodon N.E.Br., Cephalophyllum Haw. and Jordaaniella H.E.K. Hartmann. The gall midges are described from adults, pupae and larvae and information is provided on their galls, life history and distribution. Morphological attributes of the gall midges support the description of five distinct species. Furthermore, morphological characters of the Lasioptera species described here do not fit entirely with those of the genus, suggesting that a new genus should be established for them. These results provide a mere snapshot of the gall midge diversity that is to be found on Aizoaceae in southern Africa, and much work is still to be done on the Cecidomyiidae of South Africa as a whole. Further targeted sampling may reveal greater distribution ranges and additional host plants for the five species described here, as well as many more undescribed species across the Aizoaceae.

Understanding how marine biodiversity is distributed, and what drives these patterns, relies on good descriptions of marine ecosystems. This information should inform the protection of biodiversity and guide its management. Relationships between marine landscapes and biodiversity therefore need to be described at scales that are useful to regional management. Simultaneous sampling of marine biodiversity and the seafloor is challenging, so baseline ecosystem descriptions are often mismatched in their abiotic and biotic components. Cameras can sample the seafloor and its associated biodiversity concurrently, with good coverage and at low cost. These are important considerations for sustainable monitoring to inform conservation management in resource-limited regions. Terrestrial landscape characterisations cannot simply be translated to the ocean because interpreting remote ocean terrain assessments in a manner relevant to ecological analysis is complicated by depth, circulation, light attenuation, and other oceanographic variables. The integration of some of these concepts into rapid marine biodiversity assessments therefore needs ground-truthing where they are applied in new regions, to advance sustainability in long-term marine monitoring. This thesis investigated the relationship between landscape composition and benthic marine biodiversity in False Bay, South Africa using novel methods that extended biodiversity sampling across more depths and habitats than any single, previous survey of the bay. This study's approach piloted sampling and interpreting the marine landscape and biodiversity over matching spatial and temporal scales. The coverage, repeatability and ecosystem-level scale applied to this study make it a useful basis to develop monitoring protocols that are appropriate to conservation management at relevant regional scales. New insights for the region include a) a new description of the seafloor using classifications that explain the variation in epibenthic megafauna and ichthyofauna communities, b) a quantitative account of the epibenthic megafauna on the eastern reefs where species diversity was highest, and c) the synthesis of seafloor descriptions with the epibenthic megafauna and ichthyofauna to describe nine regions of False Bay, relative to two previous descriptions of "grounds". Photographs and multibeam bathymetry characterised the seafloor on eight transects across the bay and were ground-truthed by grab samples repeated at representative sites. Two new classifications were applied to describe the seafloor. Horizontal seafloor heterogeneity was highest in the east, and reef was distributed along the eastern and western margins. The Collaborative and Automated Tools for Analysis of Marine Imagery (CATAMI) scheme captured accurate broad-scale descriptions of the physical landscape when applied to photographs. Grabs are still needed to provide fine-scale particle size data on soft sediments where most invertebrate diversity is likely infauna. However, CATAMI abstracts fine-scale sediment variation into simpler groupings more useful for rapid ecosystem assessment. Photographic sampling is repeatable, which is useful for long-term ecosystem monitoring. Photographs taken using a jump camera rig assessed the epibenthic megafauna across habitats and along depth gradients. Rényi diversity showed that species diversity increased in shallow waters up to 40 m, reaching a peak between 30 and 40 m, before decreasing with increasing depth. Species diversity was highest in the east, where seafloor heterogeneity was also highest. This result is interesting because eastern False Bay falls mostly outside the current marine protected area (MPA) network and has been relatively under-represented in previous surveys. The jump camera documents ecosystem-level biodiversity patterns and processes, and the random point count method in Coral Point Count (CPCe) is useful to assess community composition and cover on reefs. The relative abundance and distribution of ichthyofauna were assessed using baited remote underwater video systems (BRUVs). Fifty-seven fish species from 30 families were recorded between 4 and 84 m. Rényi diversity showed that species richness was similar for reef and sand overall, but the Shannon-Wiener diversity index (H') was significantly higher on reef sites than on sand sites (t = 1.972, p < 0.0001). Species richness for the whole bay was similar in winter and summer, which indicates that the same species are likely present year-round; however, the Shannon-Wiener diversity index was significantly higher in winter (t = 1.973, p < 0.013) and evenness was greater in winter at the level of the site. These findings highlight the difficulty in protecting sufficient sand habitat to encompass the patchy distribution of sand-associated species and highlight seasonal differences in optimal visibility for future camera monitoring surveys by conservation management. There are clear patterns in the marine biodiversity of False Bay, at various scales, that can be detected using novel methods for the region. The study's approach to classifying both the landscape and its associated biodiversity creates a framework for future ecosystem threat assessment that can be applied elsewhere, especially along the South African coastline.

The dynamic Benguela Upwelling System is one of four major upwelling regions in the world and is subdivided into two sub-systems, the northern and southern Benguela. This current study was conducted within the southern Benguela, which lies between 27°S and 35°S (Orange River Mouth to East London) and is characterized on the west coast by seasonal, wind-driven, coastal upwelling. The study targeted three picophytoplankton groups, Synechococcus, Prochlorococcus and picoeukaryotes, which are the three most abundant < 2 µm size class phytoplankton. Flow cytometry was employed to enumerate picophytoplankton abundances, using their pigments and cell sizes to identify the different groups. The aim of the study was to investigate the role of picophytoplankton in the southern Benguela coastal upwelling ecosystem. These aims were met by: i) determining the spatiotemporal variability of each of the three picophytoplankton groups over a period of 8 sampling cruises at 44 stations across four latitudinal lines in the study region, ii) determining short-term changes in carbon and nitrogen biomass of picophytoplankton and their growth rates over a 10-day period, using abundance estimates from a station off St. Helena Bay, and iii) estimating mortality of microbial communities in a laboratory study using samples collected from a coastal upwelling environment. Results showed no strong seasonality in picophytoplankton abundances but evidence of latitudinal and zonal effects. Investigations over the short term showed that populations of picophytoplankton in the southern Benguela change on the same timescale of ~3 days as the larger phytoplankton during an upwelling event. Determining mortality rates using a dilution experiment presented some challenges. Instead of increased growth rates, the study showed decreased growth rates as predator numbers decreased. These shortcomings were investigated in a second experiment, which both excluded large predators (<200µm) and also ran a parallel experiment excluding smaller predators (10-200 µm). The last of these experiments resulted in increased growth rates as predator numbers decreased. The complexity of the southern Benguela system, with its pulsed, high productivity and large concentrations of nutrients, traditionally is known to show variability through effects on the biology of large phytoplankton. However, picophytoplankton also were variable in the study area, resulting from bottom up effects of the environment, confounded by biotic factors such as predation, parasitism and competition

Mussels living in coastal environments are often exposed to natural inorganic particles and hence may be well adapted to dealing with high sediment loads. The mechanisms by which they deal with particle loads do, however, cause stress and alter metabolic processes. An increasingly common anthropogenic addition to particle loads in the ocean are microplastic particles. Numerous recent experiments have addressed the impacts of microplastics on metabolic performance, but few of these have used natural reference particles to control for the concurrent effects of particle load itself. This study aims to compare the effects of microplastic and of natural particle exposure on the mussel Mytilus galloprovincialis, an invasive species which has become the dominant mussel in the mid- to low-shore of the south and west coasts of South Africa, but is absent from areas prone to sand inundation. These effects will be compared to those on the native mussel Choromytilus meridionalis, which resides on the low shore, and unlike M. galloprovincialis often occurs in areas prone to sand inundation. Respiration rates, byssus production, clearance rate, body condition (BCI) and survival of mussels exposed to four concentrations of two particle types, polyvinyl chloride (PVC) and red clay were measured. A significant concentration effect was found in the respiration rates of Mytilus galloprovincialis, while C. meridionalis respiration rates were largely unaffected by both particle type and particle concentration. The byssus numbers of M. galloprovincialis were significantly reduced by microplastic exposure, whilst no particle type effects were found in C. meridionalis. Clearance rates of C. meridionalis, on the other hand, were significantly affected by particle concentration, while no effects were found on M. galloprovincialis. The BCI of C. meridionalis was also found to be affected by particle concentrations, while M. galloprovincialis was unaffected. All C. meridionalisindividualssurvived the experiment, while 29 M. galloprovincialis died. Mortality of M. galloprovincialis exposed to the two particle types was not significantly different, although more mortality was suffered in PVC treatments than in red clay treatments. The results reveal that there was indeed a difference in the response of M. galloprovincialis to the different particle types, and that the two species did exhibit different strategies to both particle type, and concentration. Experimental studies of this nature are imperative in order to disentangle microplastic effects from those of particles in general, and to develop a better understanding of potential impacts of plastic debris on marine ecosystems.

This study examines the navigation of structural pressures, limitations and conservation policies by the community of Hobeni Village in the practice of traditional rituals within the context of nature conservation. It considers how ceremonies such as uKuqatywa komntwana, Intonjane, Ulwaluko, ukunikezelwa ko-Mkhontho, and Umcimbi/ Umgidi play a vital role in the construction of personhood among small groups of men, women and teenagers identifying as Xhosa people in the Hobeni Village in the Eastern Cape. The qualitative research drew insights from indigenous knowledge already known to the researcher, current research, participatory observation and semistructured interviews conducted on twenty-four members of the community. The ethnographic study found that systemic methods of nature conservation Vis a Vis symbolic oppression, sit in tension with the rituals performed by members of the AmaXhosa at Hobeni village. It argues that for Hobeni residents, accessing the natural resources placed under conservation is a vital aspect of their identity formation, which is impacted by conservation. The research found that current conservation practices pose a threat not only to AmaXhosa practices of identity formation and sense of belonging but also to the maintenance of their culture. The connection of the Hobeni people with nature is limited by conservation methods that force them to adapt their traditional practices that attempt to find congruence with their belief systems, but that strain the relationship between these villagers and their ancestors. In the search for alternative methods to preserve natural resources and maintain the culture of Hobeni village, this dissertation calls for the establishment of a different approach to conservation that is context-specific and community-centred. A transformative approach to conservation could advance environmental justice without compelling the community to negatively negotiate, as is currently the case, their cultural practices or erode their entanglement with nature. The contribution of this study lies in challenging the narrative or ideologically laden discourses that perceive people as a threat to nature and the environment. This dissertation concludes that people possess diverse knowledge systems and resources that enable them to coexist and conserve nature in their surroundings or living environment.

South African mosasaur remains consist of a frontal with articulating portions of the parietal and postorbitofrontals (SAM-PK-5265); two dentary fragments (SAM-PK-5265) assigned to ‘Tylosaurus capensis' by Broom in 1912 (SAM-PK-5265); an undescribed muzzle unit and associated isolated teeth (CGP/1/2265) from Pondoland and a more recently discovered isolated partial vertebra from St Lucia. Some research has been done on the material, but there is still uncertainty concerning their relationships and taxonomy. This research aims to provide a taxonomic assessment of all the SA mosasaurid material to better understand their phylogenetic relationships and to place them in the context of mosasaurs from other parts of Africa and globally. In addition, isotopic analyses, micro-computed tomography (micro-CT), mineralised tissue histology and scanning electron microscopy (SEM) are applied to the SA mosasaur remains to decipher various aspects of their palaeobiology. This study identifies three mosasaur taxa from SA: Mosasaurus sp., cf. M. hoffmannii., cf. Taniwhasaurus, and cf. Prognathodon. The isolated vertebra is assigned to Mosasaurus sp., cf. M. hoffmannii. The frontal and dentary fragments (SAM-PK-5265) originally described as Tylosaurus appears to be a mix of two taxa: One of the dentary fragments possesses replacement teeth with enamel ornamentation that resembles, Ta. mikasaensis, but is tentatively assigned to cf. Taniwhasaurus based on tooth recurvature. The other dentary fragment and a frontal with articulated elements are suggested to belong to the same individual as the muzzle unit for which the suggested assignment is cf. Prognathodon. Strontium analysis of tooth enamel dated the cf. Prognathodon material to the end of the Maastrichtian (87Sr/86Sr = 0.707817; age = 66.85Ma). The cf. Taniwhasaurus dentary fragment is likely Santonian-aged, as originally indicated in 1901. SEM of enamel from cf. Prognathodon reveals a complex array of prismless enamel types and pervasive aggregations of fossilised bacteria in the underlying dentine. The δ18OPO4 derived body temperature estimate (Tb) of the cf. Prognathodon (Tb = 33.21°C) compares well with previously reported Tb for mosasaurs and may indicate that the mosasaur was capable of maintaining a Tb higher than that of the surrounding seawater.

Thesium L. (Santalales: Santalaceae) is a large (360 species) genus of hemiparasitic perennial or annual species with a mainly Old-World distribution and a greatest concentration in southern Africa (ca. 186 species). Although Thesium is a major component of southern African flora, it often goes unnoticed and is poorly studied. The last revision of the entire genus was done by De Candolle in 1857. South African Thesium was last revised by Hill almost a century ago. Since Hill's revision, the number of collections have grown, and 49 new species have been described. Currently, no comprehensive Thesium taxonomic key exists, and species delimitation remains difficult due to a high variation in character states, rendering the genus in need of major revision. Within southern Africa, ca. 103 species occur in the Greater Cape Floristic Region (GCFR), of which about 72 are regional endemics. The GCFR Thesium, including ecologicalspecialists and generalists, offers an appropriate system for evaluating both the correlates of range extent, specialisation and the relative extinction risks associated with both ecological strategies. Here, it is predicted that a combination of edaphic, elevation and climate variables influence the geographic range of Thesium in the GCFR. Recent phylogenetic hypotheses revealed that Thesium is paraphyletic with respect to Austroamericium, Chrysothesium, Kunkeliella and Thesidium, suggesting the need for generic realignment. In addition, existing subgeneric and sectional classifications of this large genus lack a phylogenetic basis, thus compromising their predictive value. Using an expanded taxon sampling and a combination of nuclear (ITS) and chloroplast (matK, rpl32- trnL and trnL-F) DNA sequence data, chapter two re-assesses the phylogenetic relationships of Thesium and uses these as the basis of a new subgeneric classification of the genus. The phylogeny obtained confirms the need to place the four segregate genera into synonymy, resulting in a monophyletic Thesium. In addition, it resolves five, well-supported major clades within Thesiumwhich I recognize as subgenera. The South African endemic subgenus Hagnothesium is sister to the Eurasian subgenus Thesium (including Thesium, Kunkeliella and Mauritanica). The subgenus Psilothesium, occurring in tropical South America (formerly genus Austroamericium) and tropical Africa, is sister to the rest of the subgenera, which are all confined to South Africa. Within the latter, the subgenus Discothesium consists of subtropical and temperate species, whereas subgenus Frisea, comprising previously recognized sections Annulata, Barbata Frisea, Imberbia and Penicillata, is restricted to the GCFR. To facilitate identification of subgenera, I present identification keys, assigned species, provide brief diagnoses, identified ancestral morphological characters and, supply distribution and ecological data. Thesium subgenus Hagnothesium is endemic to the GCFR. In the past, there has been a propensity in revisionarywork ofthe subgenus Hagnothesium to split taxa into distinctspecies or vice-versa. Consequently, 15 different names exist although only six are accepted formally. Following recent molecular phylogenetic studies, the monophyly of the subgenus Hagnothesium is now well-established, but the circumscription of species within the section remains problematic given the complicated nomenclatural history which has added further confusion. Chapter three presents a revision of subgenus Hagnothesium using a total evidence approach to propose a modern taxonomy. I studied both herbarium collections and plants in their natural populations to circumscribe species boundaries, geographical ranges and estimates of their conservation status. Species of the subgenus Hagnothesium are dioecious, generally having four- merous, campanulate flowers, spikes borne in bract axils and arranged along the length of branchlets, with valvate perianth lobes and a short to absent hypanthial tube. The following eight species were recognized, of which one is here described as new: T. fragile L.f., T. fruticulosum (A.W.Hill) J.C.Manning &amp; F.Forest, T. hirtum (Sond.) Zhigila, Verboom &amp; Muasya comb. nov., T. leptostachyum A.DC., T. longicaule Zhigila, Verboom &amp; Muasya nom. nov., T. microcarpum A.DC., T. minus (A.W.Hill)J.C.Manning &amp; F.Forest and T. quartzicolum Zhigila, Verboom &amp; Muasya sp. nov. I provide updated taxonomic keys, species descriptions, illustrations, distribution maps, new combinations, synonyms, and notes on the red list status for each species. In addition, six new species of Thesium endemic to the GCFR (but not included in subgenus Hagnothesium) are described and illustrated in chapter four. These are: Thesium aspermontanum Zhigila, Verboom &amp; Muasya sp. nov., T. dmmagiae Zhigila, Verboom &amp; Muasya sp. nov., T. neoprostratum Zhigila, Verboom &amp; Muasya sp. nov., T. nigroperianthum Zhigila, Verboom &amp; Muasya sp. nov., T. rhizomatum Zhigila, Verboom &amp; Muasya sp. nov., and T. stirtonii sp. nov. Also, Thesium assimile var. pallidum is elevated to species rank as T. sawae Zhigila, Verboom &amp; Muasya stat. nov. Morphological and ecological differences between species, along with their putative affinities, preliminary conservation status, phenology, etymology and distributional maps are presented. Narrow-ranged species are expected to be more at greater risk of extinction than generalists due to climate change. Such risk is greatest in biodiversity hotspots such as the GCFR, which house both ecological specialists and generalists. It was hypothesized that range size, ecological specialization and consequent climatically-modulated extinction-risk are all phylogenetic structured, such that climate change will precipitate a disproportionate loss of phylogenetic diversity. Past and future species distribution ranges were developed using MaxEnt models based on present-day occurrences and environmental conditions. There was a strong positive correlation between the ecological niche breadth of species, as determined by large-scale environmental variables, and their range extents. One hundred and one Thesium species were modelled, of which 71 species (83%) were predicted to have had broad range sizes during the Last Glacial Maxima, and 27 species (17%) recorded range contractions historically to the present. Similarly, 45 species (44%) will potentially expand their ranges, while 51 species (50%) are predicted to reduce their ranges in the future. Of the 65 species currently ranked as Least Concern or Data Deficient in the South African Red list, 24 species will likely shift into higher extinction risk categories. Interestingly, five ecological specialists (5%), although having experienced a range reduction from the LGM to the present, are predicted to persist in the face of future climate change. However, the range extent, ecological specialisation and extinction risk are phylogenetically random and therefore should have a negligible impact on the phylogenetic diversity of the GCFR Thesium. Overall, this study confirms the monophyly of the genus Thesium and sets its infrageneric classification scheme in place. The context of this classification framework allows the systematic revision of the genus, one clade at a time. Towards this goal, I revised the Hagnothesium clade and additionally described six new species from other clades. The climate, elevation and soil variables influence the distribution range and specialism of GCFR Thesium clades. However, ecological specialism of species and extinction risks were predicted to be phylogenetically random.

Pinniped focused tourism has grown rapidly both globally and in South Africa. In a 2002 survey South Africa was identified as having earned the most revenue from seal ecotourism of all the countries that engage in this activity. Seal tourism includes approaching colonies on foot, by kayak, on large and small motorised boats and more recently in the water through snorkelling and scuba diving activities. Like most wildlife tourism seal snorkelling operators can use their tours to educate clients about seal biology and threats, while providing them with a memorable physical experience with minimal impact on the seals. The goal of this research was to explore aspects of the demography, attitudes and values of tourists participating in two different seal viewing activities at the same island. Additionally, I investigated levels of satisfaction with each tour type, differences in the style of education provided by operators and which tour provided tourists and what facts tourists found most interesting. In the second part of the study I quantified the behavioural response of seals to tourists who entered the water to snorkel with seals in an attempt to assess potential impacts of immersive trips on seal behaviour. The study was conducted at Duiker Island, near Hout Bay in the city of Cape Town between November 2019 and January 2020. This period coincides with the breeding season of Cape fur seals (Arctocephalus pusillus pusillus) and the peak tourist season. Questionnaires were distributed to both seal viewing tourists (n = 53) who had viewed seals from a large boat and seal snorkelling participants who had swum with seals at the island (n = 90). More women than men undertook both types of tour, though the proportion of men increased on snorkelling compared to viewing tours. Probit models revealed key behavioural and attitudinal differences between seal viewing and seal snorkelling participants with the latter being more pro-environmental. Seal snorkelling guides used a more interpretive style when educating tourists including the use of visual aids while seal viewing tours announced facts through a loudspeaker on the vessel. Seal snorkelling guides would also include information to awareness about plastic pollution and seal entanglement, with a donation box for a seal disentanglement program run by the Two Oceans Aquarium. Seal snorkelling participants listed an average of two facts they had learnt on the tour compared to a mean of 1.24 facts for seal viewing tours. Both tours were rated very highly for overall levels of satisfaction (seal snorkelling mean score = 9.17/10; seal viewing = 8.58/100). Surface observations of seal behaviour in response to seal snorkelers in the water close to the boat suggested a minimal impact with most seals (88%) behaving neutrally and only 0.2 % engaged in avoidance behaviour. Below water observations revealed that seal numbers declined with increasing number of people in the water and seals adjusted both their position in the water column (more diving) and their activity (more active) in response to snorkeler presence. These findings suggest that both the number of snorkelers and the area over which they spread should be controlled so that seals can choose to avoid snorkelers and behavioural changes are localised to select demarcated areas. The presence of the guides in the water together with the tourists ensured there were no inappropriate interactions between seals and snorkelers (e.g. touching and biting by seals in response). This is the first study on the thriving seal ecotourism industry at Duiker Island and reveals high levels of satisfaction by both seal viewing and seal snorkelling tourists. While there were measurable impacts of seal snorkelers on seals these were highly localised and unlikely to present an adverse impact on seals at the Island more generally. Controlling the number of operators and ensuring snorkelers are always accompanied by guides should ensure that this valuable business continues, potential impacts are minimised and negative interactions (e,g. bites from seals or people touching or feeding seals) are prevented.

This thesis investigates the evolution of Pleistocene insular proboscideans from the centralwestern Mediterranean (Palaeoloxodon species from Sicily, Malta, Favignana) and a mammoth (Mammuthus lamarmorai) from Sardinia, with a particular emphasis on the anatomy of the limbs. Differences in the morphology of the limbs are examined across a tenfold reduction in mass (from 3,5m-tall P. antiquus from Germany to 1,2 m-tall P. ex gr. P. falconeri from Spinagallo Cave, Sicily), revealing insights into significant morphological changes in the long and foot-bones, particularly appendicular changes evident in SiculoMaltese P. ex gr. P. falconeri. Notable morphological differences between P. antiquus and its insular descendent P. ex gr. P. falconeri include the functional morphology of the ankle-joint (especially the calcaneus' articular facet for the tibia). Furthermore, morphological similarities found between the femur of young continental elephants (P. antiquus and L. africana) and adult insular dwarfs (P. ex gr. P. falconeri and its probable ancestor Palaeoloxodon sp. from Lparello Fissure, Sicily) suggest evidence of paedomorphism in the limbs. Similarly, comparisons of the ontogenetic allometry of the tibia in L. africana and P. ex gr. P. falconceri include changes which are also consistent with paedomorphism, although other factors could not be ruled out. In the humerus large differences are evident in the morphology of the deltoid tubercule between co-generic insular Palaeoloxodon species, suggesting interspecific differences in the musculo-skeletal system. Furthermore, on the basis of dimensions, morphology and stratigraphy, the large Palaeoloxodon sp. remains from Luparello Fissure, north-western Sicily are suggested to belong to the ancestral chronospecies of P. ex gr. P. falconeri from Sicily, which may have subsequently colonized Malta during the reduced sea-levels of a Middle Pleistocene glacial lowstand (following a corridor with reduced distances between the two islands). Additionally, morphological differences in the calcanei of elephants from Luparello Fissure, Sicily, and Benghisa Gap, Malta may be the result of allopatric speciation between similar-sized elephants during the Middle Pleistocene, or alternatively relate to ecomorphology. These findings suggest that the morphology of the calcaneus may be more informative than hitherto recognised for resolving systematics and taxonomy among the Elephantini. Furthermore, although the absolute chronology of SiculoMaltese elephants remains poorly constrained, preliminary U-Th dating at Alcamo Quarry, western Sicily suggests a tentative early Middle Pleistocene age for Palaeoloxodon sp.