The Family Testudinidae (Suborder Cryptodira) is represented by 40 species worldwide and reaches its greatest diversity in southern Africa, where 14 species occur (33%), ten of which are endemic to the subcontinent. Despite the strong representation of terrestrial tortoise species in southern Africa, and the importance of the Karoo as a centre of endemism of these tortoise species, there is a paucity of ecological information for most tortoise species in South Africa. With chelonians being protected in < 15% of all southern African reserves it is necessary to find out more about the ecological requirements, status, population dynamics and threats faced by South African tortoise species to enable the formulation of effective conservation measures. The Leopard Tortoise (Geochelone pardalis) is the largest of the southern African species and has a wide distribution range, occurring in a variety of habitats. There is a paucity of ecological information about Leopard Tortoises in most of these habitats, especially arid and semi-arid regions. The broad aim of the study was to comprehensively investigate the ecology of Leopard Tortoises on farmland in the semi-arid Nama-Karoo biome, and use the ecological information to make recommendations for conservation. An investigation was made to determine the population size, sex and age distributions, density, biomass, and morphometrics of Leopard Tortoises in the Nama-Karoo. It was predicted that Leopard Tortoises would either be similar in size to Leopard Tortoises in other habitats or larger in order to buffer the effects of extreme temperatures by decreasing the surface-to-volume ratio. In addition, it was predicted that Leopard Tortoises would have a smaller population size, and occur at a lower density and biomass per hectare than those tortoises in mesic habitats. A total of 92 tortoises were caught, and 3899 observations made on Leopard Tortoises on 5500 hectares of farmland. Fourteen tortoises were radio tracked which allowed for repeated observations throughout the year. The population was skewed towards adults, and indicates a low hatchling recruitment. Female Leopard Tortoises were significantly larger than males with respect to all morphometric measurements. A male to female sex ratio of 1: 1.6 was obtained, which was not significantly different from 1: 1. A population estimate of 57.64 ± 3.99 tortoises for the 5500 ha area was obtained using a mark-recapture sampling method. Density of tortoises was extremely low at 0.017 tortoises.ha ¯¹, with a biomass of 0.002 kg.ha ¯¹. Population size, density, biomass and morphometrics were compared with Leopard Tortoise populations in other areas. Leopard Tortoises were larger in size in the semi-arid Nama-Karoo compared with Leopard Tortoises in other areas, probably a mechanism to reduce the effects of extreme temperature fluctuations, and were found at a much lower density. Knowledge of the home range size, home range overlap and seasonal change in home range is imperative to the understanding and conservation of the Leopard tortoise. Home range size, percentage overlap and mean daily distances moved were investigated for Leopard Tortoises as a function of season, gender and body mass. Home range and movement data were calculated for 36 Leopard Tortoises (22 females, 14 males), 14 of which were telemetered (8 females, 6 males), and 22 of which were recaptured 10 or more times (14 females and 8 males) over a period of two years. Mean (±SE) home range size for adult Leopard Tortoises was 205.41 ± 45.57 ha. Home range size was not significantly different between males and females, however females had larger home ranges than males. Mean home range size of males was 133.27 ± 32.90 ha, and of females was 251.32± 70.56 ha. There was a significant difference in home range size between telemetered tortoises (413.81 ± 89.46 ha), and those recaptured 10 or more times (72.79 ± 18.89 ha). It is suggested that unacceptable variation in home range size estimations occur when radio telemetry is not used to recapture tortoises throughout the year. No significant relationship between home range size and body mass was found for all tortoises or between sexes. Significant seasonal and gender variation existed in the seasonal home range sizes. Females had larger home ranges than males in all seasons except spring. Mean daily distance moved by Leopard Tortoises was 136.13 ± 8.94 m with males moving further overall daily than females (males: 161.10 ± 11.8 m; females: 117.07 ± 12.87), but not significantly so. Mean daily distance moved per season was significantly different between the sexes. Females covered the largest mean daily distance in autumn and males in spring. Considerable variation existed in the amount of home range overlap both within and between sexes. Overlap percentages ranged from 5% to 90%, with home ranges most frequently overlapping by 20%. Home range size and daily distances moved in the Nama-Karoo are larger than for Leopard Tortoises in other habitats. This has strong implications for the size of reserves needed in conservation efforts with regard to this and perhaps other, species in arid or semi-arid areas. Seasonal activity patterns of Leopard Tortoises were investigated as a function of rainfall, sex, time of day, temperature and time after sunrise. It was predicted that due to seasonal rainfall, and the subsequent increase in food available, the activity patterns of Leopard Tortoises would vary greatly between seasons, but that the primary constraint on activity levels within a season, would be ambient temperature. Type of activity, time of day that the activity was performed, and amount of time spent performing each activity, differed significantly between the seasons. There was no overall seasonal significant difference between the sexes and the level of activity, however, in certain seasons and with regard to specific activities, there were significant differences between the sexes. Activity patterns were primarily bimodal in summer and autumn, and unimodal in winter and spring, with non-thermoregulatory activities, for example walking and feeding, being performed primarily in the afternoon. There was a significant positive correlation between the number of tortoises caught and rainfall per season, but activity levels and the percentage of tortoises walking and feeding was not correlated with seasonal rainfall. The time of day that an activity was first performed in each season, was primarily a function of the time after sunrise and only secondarily of temperature. The response of Leopard Tortoise activity to rainfall, time of day, temperature and time after sunrise, is discussed. With Leopard Tortoises being ectotherms, they rely largely on behavioural thermoregulation to moderate the effects of daily and seasonal fluctuations in ambient temperature on body temperature. Extensive use is made of refuges to facilitate this behavioural thermoregulation. The Nama-Karoo experiences wide temperature fluctuations both daily and seasonally, and therefore the types and seasonal use of refuges by the Leopard Tortoise, in addition to the orientation of the exits and of the tortoises within the refuges, was investigated. A wide variety of refuges were used, but Lycium spp., Eberlanziaferox (Doringvygie), Opuntiajicus (American Prickly Pear) and grass clumps were preferred as refuges. There was seasonal variation in the use of these refuges that further depended on whether the refuges were used as forms or shelters. Leopard Tortoises in spring and winter often remained in the same refuge for the entire season, or returned to the same refuge on consecutive nights. There was seasonal and behavioural variation in a) compass direction that the tortoises were facing within a refuge, b) compass direction that exits of the refuges were open to, and c) portion of the shell of each Leopard Tortoise within a refuge that was exposed to sun radiation. Tortoises in winter and spring used these three factors to maximise the amount of solar radiation received on their carapace, while tortoises in summer and autumn used them to minimise solar radiation received. Therefore, using a combination of refuge type, exit orientation and tortoise orientation, Leopard Tortoises were able to passively thermoregulate and further control temperature fluctuations experienced in an extreme environment. Leopard Tortoises on farmland in the Nama-Karoo had lower densities, larger body sizes and much larger home ranges than Leopard Tortoises in other habitat types. This is an important aspect to take into account when planning for the conservation of Leopard Tortoises in semi-arid areas, and may hold further implications for other arid or semi-arid tortoise species. Activity patterns and patterns of thermoregulation allow for further understanding of the interactions between tortoises and their environment, habitat, and climate in the wild. In addition, it further aids in the understanding of the methods used by ectotherms to thermoregulate and manipulate body temperatures, especially when living in regions of unpredictable rainfall and extreme temperatures. / Thesis (M.Sc.)-University of Natal, Pietermaritzburg, 2001.