Thesis (MScAgric)--Stellenbosch University, 2014. / ENGLISH ABSTRACT: The aim of this study was to evaluate the current managerial practices as used by African Savanna buffalo (Syncerus caffer caffer) farmers. Consequently, the best management practices would be combined to formulate a basic management plan to farm captive buffalo. The distribution of buffalo throughout South Africa was also investigated and each province was considered separately for different types of buffalo (Kruger also known as project; Addo and other) and different disease statuses (Foot and Mouth; TB; Corridor disease and disease-free or clean). The basic infrastructure of all farms studied was noted and evaluated to attain the most effective structures and layouts needed for basic captive buffalo farming. The reproductive capabilities of buffalo were assessed on different farms. These farms were divided into winter and summer rainfall areas to ascertain whether season or rainfall would have an effect on calving season. Additionally the reproduction data was analysed to set a benchmark for the reproductive performance of buffalo in herds as well as individually. This assisted in selection in captive breeding of buffalo. Buffalo are currently distributed throughout South Africa and occur in all nine provinces, with the highest quantity found in Limpopo with 1300 registered buffalo farms. Provinces that contain only disease-free buffalo include Western Cape, Eastern Cape, Freestate, North-West and Gauteng. Corridor infected buffalo are found in the Northern Cape, Mpumalanga and KwaZulu Natal. Foot and Mouth disease is found in Limpopo and Mpumalanga and TB infected buffalo are found in Mpumalanga and KwaZulu Natal. Factors to consider when managing captive buffalo herds are the herd dynamics and composition, feeding and nutrition and lastly parasite control. Management should be approached adaptively as different areas present different challenges. Infrastructure is divided into the farm and biomes thereof, feeding and parasite treatment. As with herd management these should be approached adaptively as the composition of each farm differs. Reproductive maturity of buffalo is reached between the ages of two and six years. Average intercalving period of captive buffalo was to be 443 days with optimal intercalving being below 400 days. Seasonal calving differences between summer and winter rainfall areas were found with calving peaks differing by two months between these areas. / AFRIKAANSE OPSOMMING: Die doel van hierdie studie was om die bestuurstegnieke wat tans deur Afrika Savanna buffel (Syncerus caffer caffer) boere gebruik word te evalueer. Gevolglik sal die beste bestuurs praktyke gekombineer word om ‘n basiese bestuursplan the formuleer om omheinde buffels te boer. Die verspreiding van buffels in Suid-Afrika is ook ondersoek en elke provinsie is afsonderlik oorweeg vir die verskillende tipes buffels (Kruger ook bekend as projek; Addo en ander) en verskillende siektestatusse (Bek-en-Klou seer; TB; Corridor siekte en siekte-vrye of skoon). Die basiese infrastruktuur van al die plase in die studie is genoteer en geivalueer op die mees effektiewe strukture en uitlegte vas te stel wat benodig word vir die boer van omheinde buffels. Die reproduktiewe vaardighede van buffels is geassesseer op verskillende plase wat verdeel is in winter en somer reënval streke om vas te stel of seisoen of reënval ‘n invloed het op kalf seisoen. Die reproduksie data is ook geanaliseer om ‘n riglyn te stel vir die reprodutiewe prestasie van buffels in ‘n kudde asook individueel. Dit sal help met die seleksie van teel diere. Buffels is tans wyd versprei oor Suid-Afrika and kom in al nege provinsies voor met die hoogste hoeveelheid in Limpopo (1300 geregistreerde buffelplase). Die provinsies wat slegs siekte-vrye buffels bevat is Wes-Kaap; Oos-Kaap; Vrystaat; Noord-Wes en Gauteng. Corridor-besmette buffels kom voor in Noord-Kaap; Mpumalanga en KwaZulu Natal. Bek-en- Klou seer kom voor in Limpopo en Mpumalanga en TB kom voor in Mpumalanga en Kwa- Zulu Natal. Faktore wat oorweeg moet word met die bestuur van omheinde buffeltroppe is kudde dinamika en samestelling, voeding en laastens parasiet beheer. Buffelbestuur moet aanpasbaar wees aangesien verskillende areas verskillende uitdagings bied. Infrastruktuur kan opgedeel word in die plaas en sy biome, voeding en parasiet toediening. Soos met kudde bestuur moet infrastruktuur ook aanpasbaar wees, aangesien die samestelling van elke plaas verskil. Reproduktiewe volwassenheid van buffels word bereik tussen die ouderdomme van twee en ses jaar. Gemiddelde interkalf periode vir omheinde buffels was 443 dae met optimale interkalwing van minder as 400 dae. Seisoenale kalwingsverskille tussen somer en winter reënvalstreke is opgemerk met kalf pieke wat verskil met twee maande tussen die streke.

Thesis (doctoral)--Swedish University of Agricultural Sciences, 2003. / Appendix includes reprints of five manuscripts, four co-authored with others. Includes bibliographical references. Also partially issued electronically via World Wide Web in PDF format; online version lacks appendix.

Thesis (Ph. D.)--University of Pretoria, 2005. / Title from PDF t.p. (viewed Mar. 27, 2006). Includes bibliographical references.

The interactions between vegetation and climate are complex and critical to our ability to predict and mitigate climate change. Savanna ecosystems, unique in their structure and composition, are particularly dynamic and their carbon cycling has been identified as highly significant to the global carbon budget. Understanding the responses of these dynamic ecosystems to environmental conditions is therefore central to both ecosystem management and scientific knowledge. Longleaf pine ecosystems are highly biodiverse and unique savanna ecosystems located in the south-eastern USA – an important current carbon sink and key area identified for future carbon sequestration. These ecosystems depend on fire to maintain their structure and function, and the longleaf pine tree itself (Pinus palustris Mill.) has been noted for its resilience to drought, fire, pests and storms and is thus becoming increasingly attractive as both a commercial forestry species and a provider of other ecosystem services. Previous process-based models tested in the south-eastern USA have been shown to fail in conditions of drought or rapid disturbance. Consequently, in order to inform management and understand better the physiology of these ecosystems, there is a need for a process-based model capable of upscaling leaf-level processes to the stand scale to predict GPP of longleaf pine savannas. P.palustris exists across a wide range of soil moisture conditions, from dry sandy well-drained soils (xeric) to claypan areas with higher moisture content (mesic). Previous work has demonstrated that this species adjusts many aspects of its physiology in response to these differing soil conditions, even under identical climate. The research in this thesis supports these previous findings, but additionally explores, with the assistance of the Soil Plant Atmosphere model (SPA), the productivity response of P. palustris across the soil moisture gradient. Contrary to expectations, measurements, field observations and modelling suggest that P. palustris trees growing in already water-limited conditions cope better with exceptional drought than their mesic counterparts. At the leaf-level, xeric P. palustris trees were found to have higher measured net photosynthesis, but the lower stand density and leaf area at this site meant that in non-drought conditions mesic P. palustris annual gross primary productivity (GPP) was 23% greater than xeric annual GPP. Initial upscaling of leaf-level processes to the canopy scale using the SPA model found that, during the growing season when other components of longleaf pine ecosystems are active, the longleaf pine may only be responsible for around 65% of the total productivity. Other important components of longleaf pine savannas are oaks and grasses which, with pine, constitute 95% of longleaf pine ecosystem biomass. Each of these groups, however, responds differently to fire and water availability. Despite this, the other components of longleaf pine savannas have received limited research attention and have never been modelled using a process-based model such as SPA. As integral components of longleaf pine carbon budgets, it is essential that the physiology and productivity of oaks and grasses in this system are better understood. The research in this thesis studied the productivity response of these groups during drought across a soil moisture gradient, and found that oak and pines at each site appear to fill separate ecohydrological niches depending on whether or not they are growing in a xeric or mesic habitat. As expected, the highest drought tolerance was found in the C4 grass, wiregrass (Aristida stricta), at both xeric and mesic sites. In order to further explore the contributions of the different functional groups in longleaf pine savannas, the SPA model was adapted to run with concurrent functional groups and to represent the different photosynthetic pathways of the understorey grasses (C4) and the canopy trees (C3). The aim of this part of the thesis was to represent better a savanna ecosystem in a process-based model and explore and quantify the contributions of each functional group diurnally, seasonally, annually and interannually. Modelling results suggest that accurately representing the phenology not only of trees but of grasses, is critical to capturing ecosystem GPP and its variability. This phenology may not only be seasonally controlled, but also dictated by fire. Overall, this research highlights the importance of continued research into savanna and savanna-like ecosystems. Additionally, it provides an insight into the responses of multiple ecosystem components to an extreme drought, and how these responses differ at leaf, stand and landscape scales. The thesis also employs a little-used method of combining eddy-covariance data with a process-based model to separate out different ecosystem components, a method becoming more common but not yet widely tested.

There has been a critical decline in grassland bird populations due to habitat fragmentation and deterioration, and suppression of natural fires. Alteration of the disturbance cycle may lead to changes in vegetation structure and thus habitat suitability for breeding grassland birds. Management practices at the Mississippi Sandhill Crane National Wildlife Refuge, including the use of prescribed fire, are in need of evaluation. My study asked what frequency of prescribed burns is necessary to support breeding grassland birds and whether vegetation structure varies among burn units. In this study, bird abundance and species richness did not differ significantly among burn units and vegetation cover-type was not a strong predictor of these factors either. There was evidence of site utilization by breeding grassland birds immediately following a burn, which suggests that the bird community is able to recovery quickly post-fire and these units may serve as viable habitat for breeding grassland birds.

A dissertation submitted to the Faculty of Science, University of the Witwatersrand, in fulfilment of the requirements for the degree of Master of Science 17th August 2015 in Johannesburg, South Africa / Ecosystem services provide a vital lifeline to millions of people living in rural areas. The poorest people in these areas depend upon the natural resource base in their surroundings to provide these services. With growing populations in rural areas of South Africa, the natural resource base is under considerable pressure; however, uncovering the dynamics of vegetation in these systems has proven difficult. While much attention has been given to savanna ecology, long term studies on the patterns and drivers of woody biomass are few. We used 65 years of aerial imagery (from 1944 to 2009) over 31 953 ha of rural savanna in a communal rangeland in South Africa to determine the abundance of woody canopy cover. This data were captured at hectare resolution, giving a fine enough level of detail for local level analysis. We also captured data for five potential drivers for change at this resolution, in order to analyse these drivers for their relative importance in determining woody canopy cover throughout the study period. Surprisingly, while individual sites showed varied trends in the amounts of woody canopy cover through time, when pooled across all sites the total woody canopy cover increased over the 65 year period. Disturbance gradients were found around some of the villages, but only in 2009, suggesting that the drivers of disturbance gradients in these systems may have only operated sufficiently to produce disturbance gradients in recent years. A hot spot analysis (hot spots indicate cells that have similarly high values beyond what would be expected in a random distribution, with cold spots indicating the inverse) revealed an increase in both hot and cold spots through time, but with a low persistence of both through time. High canopy cover cells are presumed to be the result of bush encroachment, while low canopy cover cells are presumed to be the result of harvesting of trees for fuelwood or clearing for fields. The low persistence of hot and cold spots points to a system in continual change, with patches of hot and cold spots appearing and disappearing, and therefore drivers of change operating in short periods of time. MAP (Mean Annual Precipitation), and not an anthropogenic driver, was found to be the most important driver for woody canopy cover throughout the study period, with MAP up to 670 mm having a predictable pattern of hot and cold spots through time. Higher MAP was shown to have a non-linear and unpredictable pattern of hot and cold spots through time, indicating that low precipitation may produce a system where woody canopy cover is less influenced by other drivers and is more stable when acted upon by other drivers. This research demonstrates the value of a long term dataset, and the applicability of our methods for monitoring woody canopy cover. As such, it may well serve as a baseline for woody canopy cover in communal savanna rangeland systems, with the methodology employed here suitable for an early warning detection system for sudden changes in the woody canopy cover.

A dissertation submitted to the Faculty of Science, University of the Witwatersrand, Johannesburg, in fulfilment of the requirements for the degree of Master of Science Johannesburg, September 2015 / Plant functional traits provide a means to investigate the diverse ecological strategies employed by plants and a tangible link to assess how the variability in these traits might influence ecosystem processes and functioning. The aim of this dissertation has been to determine how plant and soil nutrient dynamics in a savanna environment are affected by two primary drivers, one a top-down driver, being herbivory by large mammalian herbivores and the other a bottom-up driver, the variable N2-fixation capacity of tree species. To the best of my knowledge this is the most comprehensive study to date to investigate the bioavailability of soil nutrients and the link between these availabilities and plant functional traits. Furthermore this study provides important insight into the use of a novel technology, ion exchange resin capsules in a South African savanna context. By studying a selection of plant functional traits (nutrient concentrations, ratios as well as specific leaf area, relative chlorophyll content and leaf dry matter content) and soil nutrients (suite of macro- and micronutrients) associated with two species of savanna tree of contrasting N2-fixation capacities, I went about investigating how herbivory differentially influences the nutrient dynamics of this system. Selecting individuals of the N2-fixing Acacia tortilis and the non-N2-fixing Combretum hereroense both inside an exclosure and on the adjacent land allowed me to determine the potential impacts by herbivores. These include both direct impacts from foraging and indirect impacts through the regulation of nutrient input pathways via deposition of dung and urine. The work compiled for this dissertation is based on the experimental work conducted in a mesic savanna system in the Marakele Park (PTY) Ltd. During the course of this dissertation, I investigated herbaceous and woody biomass in relation to protection from and exposure to herbivory, determining any differences in the functional leaf traits between individuals inside and outside the exclosure, if these differences were exhibited in the associated herbaceous biomass as well as a comprehensive assessment of the bioavailability of 15 important micro- and macronutrients using ion exchange resin capsules. These capsules were incubated in the soil over the entire summer rainfall period, providing a cumulative view of nutrient bioavailability during the growing season. In this work I also demonstrated whether particular nutrients are associated with specific drivers (i.e. herbivory, canopy position or N2-fixation). Furthermore, these results were then looked at together to suggest the mechanism by which herbivory and N2-fixation drive nutrient dynamics and make recommendations on the use of these results in managing savanna systems in the future. Between the two sites, aboveground herbaceous biomass was significantly greater when protected from herbivores than on the adjacent land. Both exposure to herbivory and N2-fixation capacity were found to alter plant functional traits. Herbivore presence was associated with an increase in herbivore-resistant or structural traits such as C/N, C/P, foliar C and SLA as well as a reduction in N and P content. These less palatable leaves were accompanied by a significantly lower availability of a number of important soil elements, namely NO3-N, inorganic N, P, K, Na, Cu, B, Mg, and S. This suggests a feedback loop between these two components of the ecosystem. N2-fixation capacity is associated with greater concentrations of elements such as N and P and a reduction in traits that are illustrative of a greater structural investment into leaves. Soil nutrient bioavailability however, shows a reduction in certain nutrients when associated with Acacia. A number of nutrients which show a reduction in availability are those which are essential to N2-fixation machinery, namely B and Fe but also lower bioavailabilities of Al and Mg. Finally, Ca, NO3-N, B, Fe, Al and inorganic N were found in greater quantities below the tree canopy than beyond it. In conclusion both herbivory by large mammalian herbivores and N2-fixation have significant effects on tree health, through their regulation of limiting nutrients and alteration of leaf traits. Given the changes which these drivers are capable of exerting on plant and soil nutrient dynamics, this has important consequences for ecosystem processes and functioning and highlights potential considerations in the long-term sustainable management of savannas.

O Cerrado é o bioma brasileiro que mais perdeu área para o agronegócio em anos recentes, verificando que a expansão da fronteira agrícola sobre o Cerrado provocou a pulverização dos remanescentes em fragmentos cada vez menores, imersos em matrizes de baixa similaridade, causando uma série de impactos negativos sobre o ecossistema natural. A fragmentação de habitats é considerada uma das mais graves ameaças à conservação da biodiversidade, com consequências ecológicas variáveis, intensificadas na faixa de transição entre matriz e fragmento. Os efeitos de borda são decorrentes de mudanças físicas e bióticas nas faixas marginais das ilhas de vegetação natural, que promovem modificações nas características microclimáticas, as quais conduzem a uma série de alterações importantes nas comunidades que compõem os ecossistemas. O objetivo deste estudo foi investigar a existência de efeito de borda e suas consequências sobre a comunidade vegetal do Cerrado stricto sensu. O estudo foi realizado em fragmento de Cerrado com área de 980,8 ha, localizado no município de Iaras, SP, cujo entorno foi ocupado por pastagem durante cerca de um século. Efetuou-se a caracterização estrutural e florística de diferentes estratos da vegetação em diferentes distâncias da borda (0, 10, 20, 40, 80 m) e coletaram-se dados microclimáticos na matriz (a 10 m de distância da borda) e no interior do fragmento, nas diferentes distâncias da borda e em duas estações do ano (seca e chuvosa). Investigou-se a existência de um possível gradiente de estrutura ou composição da vegetação relacionado com a distância da borda e com as variáveis microclimáticas. Os resultados obtidos mostraram não existir variações de microclima e de estrutura e composição do componente arbóreo e herbáceo-arbustivo em relação à distância da borda do fragmento. Embora independente do microclima e, portanto, diferente dos clássicos efeitos de borda, a invasão pela braquiária (Urochloa decumbens) foi detectada como a única consequência da exposição da periferia do Cerrado stricto sensu às pressões da matriz, causando exclusão das gramíneas nativas e inibindo o desenvolvimento de plantas lenhosas de pequeno porte. A ausência de efeito de borda sobre os demais componentes da vegetação sugere, em um primeiro momento, que esta não é uma ameaça importante para a conservação do cerrado. Porém, a invasão biológica pela gramínea africana é relativamente recente (foi introduzida há cerca de três décadas na propriedade) e suas consequências ainda não podem ser avaliadas adequadamente. É possível que seus impactos sobre os estratos superiores da vegetação venham a ser observados no futuro, quando a inibição da regeneração das espécies arbustivas e arbóreas comprometer a dinâmica da comunidade vegetal e a substituição dos adultos nas populações dessas espécies. Além disso, hoje a invasão só causa impactos significativos sobre a faixa de 0 a 20 m de distância da borda, mas não é possível prever se a gramínea invasora vai se expandir rumo ao núcleo do fragmento no futuro. / From the Brazilian biomes, the Cerrado has been the most widely and rapidly converted to agribusiness in recent years. The expansion of agricultural frontier over the Cerrado domain resulted in thousands of small fragments embedded in matrices of low similarity, causing a series of negative impacts on the natural ecosystem. Habitat fragmentation is considered one of the most serious threats to the conservation of biodiversity and ecological consequences vary, being intensified in the range of transition between matrix and fragment. The edge effects are caused by biotic and physical changes in marginal areas of the islands of natural vegetation, promoting changes in microclimatic characteristics, which leads to a series of important changes in the ecosystems. The aim of this study was to investigate the existence of edge effect and their consequences on the plant community of Cerrado sensu stricto. The study was conducted in Cerrado fragment with an area of 980.8 ha, located in the municipality of Iaras, SP, whose surrounding area was occupied by pasture for about a century. Structure and floristic composition of distinct vegetation layers was assessed at different distances from the edge (0, 10, 20, 40 and 80 m). Climatic data were collected in two seasons (dry and wet), at the same distances withing the fragment and also in the matrix (10 m away from the edge). The existence of a gradient of vegetation composition or structure and the microclimatic variables related to distance from the edge was investigated. No changes in microclimate and structure or composition of the wood layers related to distance from the edge was observed. Although independent of the microclimate and therefore different from classic edge effects, invasion by the African grass Urochloa decumbens (braquiária) was detected as the only consequence of exposure of the periphery of the cerrado sensu stricto to the pressures of the matrix, causing the exclusion of native grasses and inhibiting the development of small woody plants. The absence of edge effect on other components of the vegetation suggests, at first, that this is not a major threat to the conservation of the Cerrado. However, the biological invasion by the African grass is relatively recent (it was locally introduced about three decades ago) and its consequences have not been adequately assessed. It is possible that their impacts on the upper strata of the vegetation may be observed in future, when the inhibition of regeneration of shrub and tree species jeopardize the dynamics of plant community and the replacement of adults in the populations of these species. In addition, at present the invasion has significant impacts only over 0-20 m from the edge, but it is impossible to predict whether the invasive grass will expand or not into the core area of the fragment in the future.

No presente trabalho, estudamos uma área de vegetação natural, composta principalmente por cerrado, com 1269 ha, situada no munícipio de Santa Rita do Passa Quatro, estado de São Paulo (21 36-44 S e 47 34-41 W). O levantamento florístico da área foi efetuado em 18 excursões de coleta mensais, com duração de 3 ou 4 dias cada, entre setembro de 1995 e fevereiro de 1997. Em cada uma das formações e fisionomias de cerrado existentes (cerradão, cerrado sensu stricto, campo cerrado, campo úmido, mata ciliar, floresta estacional semidecídua), o material em fase florífera ou frutífera ou de formação de esporos foi coletado e, posteriormente, prensado e identificado em nível específico. Coletamos 1944 exsicatas, representando 499 espécies, 317 gêneros e 107 famílias. As famílias que se destacaram foram: Asteraceae, Fabaceae, Poaceae e Rubiaceae. A comparação das formações e fisionomias de cerrado através de índice de similaridade ressaltou a unidade florística do cerrado. Para auxiliar a identificação das espécies coletadas, construimos chaves baseadas em caracteres vegetativos para as formações de cerrado sensu lato, campo úmido, mata ciliar e floresta estacional semidecídua. Para estudar a estrutura da comunidade vegetal, lançamos aleatoriamente parcelas nas fisionomias de cerrado e na floresta estacional semidecídua, cujos tamanhos e critérios de inclusão foram diferentes em função de variações no componente lenhoso. A biomassa, estimada pelo volume cilíndrico total, aumentou do campo cerrado à floresta estacional semidecídua, com valores intermediários no cerrado sensu stricto e no cerradão. O cerrado, em suas três fisionomias, apresentou não só grande diversidade alfa, como também altas diversidades beta e gama. A partir dos dados do levantamento florístico, estudamos as variações fenológicas das espécies, procurando analisá-las como estratégias adaptativas. A proporção de espécies anemo e autocóricas foi maior no componente herbáceo-subarbustivo, ao contrário das zoocóricas, mais freqüentes no componente arbustivo-arbóreo. As espécies arbustivo-arbóreas floresceram principalmente no início da estação chuvosa, enquanto que as herbáceo-subarbustivas produziram flores, de modo geral, apenas no final da estação úmida, após período de acúmulo de carboidratos. A proporção de espécies anemo e autocóricas em frutificação foi maior na estação seca, quando sua dispersão é mais eficiente. Já as espécies zoocóricas frutificaram principalmente durante toda a estação quente e chuvosa, quando seus frutos carnosos podem se manter atraentes por mais tempo. / We studied a natural vegetation area, composed mainly by cerrado, with 1269 ha, located at Santa Rita do Passa Quatro municipality, São Paulo State, southeastern Brazil (21 36-44 S e 47 34-41 W). We carried out a floristic survey in 18 field trips, each one with 3 or 4 days, from September 1995 to February 1997. On each existing vegetation types (three cerrado physiognomies, floodplain grassland, gallery forest, and seasonal semidecidous forest), all vascular plants in reproductive stage were collected and then pressed and identified in species level. A total of 499 species, representing 317 genera and 107 families, was found. The most important families were: Asteraceae, Fabaceae, Poaceae and Rubiaceae. The comparison of all vegetation types and cerrado physiognomies through similarity index emphasized the cerrado floristic unity. To help the species identification, we constructed four indented keys based on vegetative characters, one for each vegetation type (cerrado, floodplain grassland, gallery forest, and seasonal semidecidous forest). To study the plant community structure, we placed randomly quadrats in the cerrado physiognomies and in the seasonal semidecidous forest, with different sizes and inclusion criteria according to its woody component. Biomass, estimated by total cylindric volume, increased from campo cerrado to seasonal semidecidous forest, with intermediate values in cerrado sensu stricto and cerradão. Cerrado, on its three physiognomies, showed high alpha diversity as well as beta and gamma ones. Based on the data collected during the floristic survey, we studied the phenological variations, analysing them as adaptive strategies. The ratio of anemo and autochorous species was greater in the herbaceous component. The zoochorous ones, on the other hand, were more frequent in the woody component. The woody species flowered mainly at the beginning of the rainy season, while the herbaceous ones produced flowers generally at the end of that season, after a period of carbohydrate accumulation. The anemo and autochorous species produced fruits principally at the dry season, when its dispersion is more efficient. The zoochorous ones fruited along the whole rainy and warm season, when its fruits become attractive for longer time.

A distribuição e a abundância das espécies de lepidópteros podem ser determinadas por vários fatores tais como disponibilidade de recursos, fatores genéticos, climáticos e inimigos naturais. Este estudo avaliou as relações entre espécies de Lepidoptera e de Malpighiaceae encontradas em dois fragmentos de cerrado do Estado de São Paulo. Para tanto, o trabalho procurou responder às questões: 1) Qual a similaridade da fauna de lagartas entre as espécies de Malpighiaceae?; 2) Quais as características vegetais que influenciam a diversidade de lepidópteros? 3) As mesmas relações se mantêm em fragmentos distintos? Os resultados encontrados demonstram que espécies de mesmo gênero de Malpighiaceae não apresentam necessariamente maior similaridade de fauna de lepidópteros, uma vez que houve maior similaridade entre Byrsonima intermedia, Byrsonima coccolobifolia e Banisteriopsis pubipetala. Não houve correlação direta entre algumas características vegetais como: a concentração de nitrogênio e taninos, a densidade de tricomas e a herbivoria. As defesas vegetais das Malpighiaceae parecem influenciar a diversidade destes herbívoros, uma vez que a riqueza e abundância estão relacionadas à essas características. A presença de alcalóides e a fauna associada a Banisteriopsis stellaris reforça essa sugestão. A similaridade da fauna de lagartas entre os dois fragmentos de 42,2% e algumas diferenças observadas nas concentrações de nitrogênio e taninos nas mesmas espécies vegetais nas duas áreas sugerem que fatores ambientais influenciam o fenótipo das plantas e as associações com seus herbívoros. Com base nos resultados encontrados, são sugeridos estudos complementares para melhor compreensão das complexas relações entre lepidópteros e as espécies de Malpighiaceae. / Species distribution and abundance are determined by several factors such as resource availability, genetics, weather and natural enemies. The present study focused on the relationships between Lepidoptera and Malpighiaceae species on two Cerrado fragments at São Paulo State. The following questions were addressed: 1) How similar is the caterpillar fauna among Malpighiaceae species?; 2) What are the plant characteristics influencing Lepidoptera diversity? 3) Are plant-Lepidoptera relationships kept on distinct fragments? Present data suggests that plants of different species of same genus have not, necessarily, the highest faunal similarity, e.g., the highest similarity among Byrsonima intermedia, Byrsonima coccolobifolia and Banisteriopsis pubipetala. No direct correlations were found among some plant characteristics such as nitrogen and tannin concentrations, trichome density and herbivory. However, Malpighiaceae defenses probably influenced herbivore diversity, because Lepidoptera species´ richness and abundance were related to these characteristics. Alkaloids and the specific fauna of Banisteriopsis stellaris support this suggestion. The caterpillar fauna similarity of 42,2% between fragments and some differences on nitrogen and tannin concentrations of same species from distinct fragments suggest that environmental factors affect, considerably, plant phenotype and plantherbivore interactions. Based on these results, some complementary studies were suggested to approach the complex relations between Lepidoptera and Malpighiaceae species.