Host suitability is critically important to the success of brood parasitism. Parasites must select a host that not only accepts its egg but also is capable of successfully rearing the parasite to fledging. Nearly all brood-parasites appear to avoid low-quality hosts that are likely to reject their eggs, that are of inappropriate size, or that feed their nestlings nutritionally inadequate or insufficient food. The diderik cuckoo, (Chrysococcyx caprius), is an obligate brood parasite known to parasitise a wide spectrum of ploceids, including the yellow weaver (Ploceus subaureus) and the southern red bishop (Euplectes orix). Theory predicts that brood parasites should exploit insectivorous passerines of similar adult size to themselves that provision large quantities of high protein food to their young. However, the relatively smaller granivorous red bishop is the most heavily parasitised host species of the diderik cuckoo in southern Africa. To investigate why an apparently unsuitable host species was so heavily parasitized several populations of parasitised red bishops and yellow weavers (omnivores) were studied in the Pietermaritzburg region, KwaZulu-Natal, South Africa. Host suitability was assessed by examining diet quality, host-provisioning rates, cuckoo nestling growth and cuckoo fledging success. Diderik cuckoo nestlings were provisioned the same diet as the host nestlings in red bishop and yellow weaver nests. However, cuckoos in bishop nests received a protein-deficient seed diet from as early as six days following hatching. In contrast, weaver-cuckoo faeces contained 1~ times more insect than their bishop counterparts throughout their nestling period. Provisioning rates by bishop females were significantly slower than by yellow weavers, and neither bishop nor weaver hosts showed any 'supernormal' effort when feeding a young cuckoo. Both host species provisioned cuckoo nestlings at a similar rate and with a similar food mass as their own nestlings. Diderik cuckoos in bishop nests grew at a slower rate and fledged in a poorer condition than their yellow weaver counterparts. Red bishops are likely the most exploited host of the diderik cuckoo because i) cuckoo eggs are more readily accepted by the less discriminating bishop and ii) the bishop-breeding season coincides more closely with that ofthe diderik cuckoo than the yellow weaver. Thus, diderik cuckoos may preferentially exploit bishop hosts because of the low frequency of cuckoo egg rejection, which ultimately results in many cuckoos fledging from bishop nests despite the lowquality diet provisioned and 53% (n = 53) fledging success in nests of this species. In the yellow weaver system, the protein-rich diet and the greater probability of cuckoo nestling survival (80%, n = 5) may compensate for the high rejection rate of cuckoo eggs by this host. Thus, both host systems seem to represent evolutionary compromises for the diderik cuckoo, with neither red bishops nor yellow weavers being entirely ideal as host species. / Thesis (M.Sc.)-University of Natal, Pietermaritzburg, 2003.

The by-catch of penaeid prawn trawlers operating on the Tugela Bank of Natal was analysed at irregular intervals for a period of three years, from May 1989 to June 1992. This was primarily in response to claims by Iinefishermen that trawlers were catching and discarding juvenile Iinefish species. Catch and effort data were recorded from 159 trawls, and catch composition data from 110 trawls, over a depth range of 20 to 45 meters. A total of 108 teleost species were identified in the by-catch, six of which comprised 80 percent of overall teleost abundance by number. Many of these fish were small-sized shoaling species, which increased their vulnerability to trawlers, and the majority were considered to be estuarine-associated. Multivariate analysis of samples showed there to be significant differences in relative abundance of species by season and depth. Annual average by-catch and discarded quantities were estimated at 400 and 315 tonnes respectively. Discard:retained catch ratios' ranged from 4.25 (± 1.19) : 1 to 15.48 (± 3.86) : 1 for shallow (20-33 meters) and deep (33-45 meters) trawls respectively. Only one important Iinefish, the squaretail kob, Argyrosomus thorpei, occurred in prawn trawl catches. These fish were spawned four to six months prior to being caught and appeared to use the Tugela Bank as a nursery area. They were seasonally abundant, occurring mostly from January to March. During these months trawling effort was generally low, as the prawns had not yet migrated offshore in quantity. Per-recruit analyses showed that the removal of these age 0 fish by trawlers reduced yield-per-recruit and spawning biomass-per-recruit values for the A. thorpei fishery. Aspects of the biology of three common sciaenids occurring in trawl catches, Johnius dussumieri, J. amblycephalus and Otolithes ruber, were investigated. Most of the fish examined were juveniles which utilised the Tugela Bank as a nursery area, owing to the existence of suitable food, and the turbid waters afforded them protection from predators. Standard biological data, such as length frequencies, lengthweight relationships and size at 50% maturity are presented. Based on the data collected and the analyses presented here, management recommendations are presented, which include the improvement of the collection of catch and effort data for the Tugela Bank prawn trawlers; communication of the results of this study to prawn trawler operators in order to encourage the reduction of trawling effort early in the year; consideration to be given to the re-surveying of Tugela Bank ichthyofaunal by-catch in the future in order to assess potential long-term effects of trawling in the area. / Thesis (M.Sc.)-University of Natal, 1993.

Mitochondrial DNA (mtDNA) was isolated from 8 different natural populations of the rodent species Otomys irroratus (Muridae: Otomyinae) and from one population of the species 0. angoniensis occurring in South Africa. MtDNA samples were cleaved with five different restriction endonucleases, end-labelled with phosphorous-32, separated by electrophoresis on horizontal 1 % agarose gels and the resulting fragment bands were detected by autoradiography. The individual-specific fragment banding patterns were analysed and compared among the various populations. The percent sequence divergence among and between the populations was calculated using the formula of Nei (1979). A matrix of sequence divergence values for all intergenomic pairwise comparisons was subjected to a clustering analysis by the unweighted pair group method with arithmetic means (UPGMA, Sneath and Sokal, 1973), using the computer programme NTSYS (Rohlf: 1988). The results of these analyses allowed for a preliminary identification of phenetic groupings in the data set. A matrix generated by scoring the restriction endonuclease fragments as present or absent was used to generate a phylogenetic dendogram using the BIOSYS (Swofford and Selander, 1989) programme. The overall restriction fragment variation uncovered in this study revealed 15 different mtDNA haplotypes within the 20 individuals examined. This corresponded to a high degree of polymorphism in the populations where more than one specimen was available, as well as within the species 0. irroratus. There were no clones that were shared between any of the populations sampled. The intrapopulation sequence divergence values uncovered in this study were high (range 0.35 % to 5.08 %), but also consistent with some other reports in the literature for intrapopulation variation. The outgroup, 0. angoniensis revealed the highest divergence values when compared to the mtDNA clones found in 0. irroratus. The phenetic and cladistic cluster diagrams revealed overall similarity with one another. There appeared to be little correlation between the topology of the mtDNA haplotype phenograms and the geographic distance of the sample localities. There was, however, a marked congruence between the distribution of mtDNA haplotypes and the distribution of three distinct cytotypes occurring over the species range. A possible polyphyletic evolution of populations of 0. irroratus was inferred from the cladistic analysis. / Thesis (M.Sc.)-University of Natal, 1994.

An interpopulation study was done on the rodent species Otamys irroratus (Muridae:Otomyinae) using restriction fragment length Polymorphisms to examine the mitochondrial DNA (mtDNA) of 30 vlei rats (Otamys irroratus) from three South African locations and 12 Angoni vlei rats (O.angoniensis) from two locations which were included as an outgroup. The three O.irroratus Populations originated from Karkloof and Kamberg in the Natal midlands and from Rietvlei in the Southern Transvaal. Mitochondrial DNA was extracted and purified by cesium-chloride/ethidium-bromide ultracentrifugation and digested with 19 class 11 restriction endonucleases. The fragments were end-labelled with 32P-dCTP, separated by electrophoresis on horizontal 1% agarose gels and the bands detected by autoradiography. The resultant individual-specific fragment patterns were analysed using the Restsite analysis program (v 1.1; Nei and Miller, 1990) to obtain a measure of the percent sequence divergences between and within the 3 POpulations of O.irroratus as well as between this species and the outgroup. The 19 endonucleases detected 19 distinct O.irroratus mtDNA maternal lineages and 3 O.angoniensis lineages. The O.irroratus lineages were clearly geographical ly structured and most closely reflected the Avise et al. (1987) category I (phylogenetic discontinuity with spatial separation). The only exception was a possible ancestral lineage represented by single individuals from Kamberg and Karkloof. Phylogenetic affinities between the most diverse lineages found at Kamberg and most Karkloof clones appear to be consistent with the finding of Pillay et al. (1993) and Contrafatto et al. (1992b) that Kamberg O.irroratus is an incipient sibling species of Karkloof O.irroratus. The mtDNA data indicates that the O.irroratus Populations at Karkloof and Kamberg last shared a common ancestor approximately 365 000 years ago. By contrast, O.angoniensis showed no evidence of geographic mtDNA structuring and is best described by the Avise et al. (1987) category Ill, which reflects phylogenetic continuity with spatial separation. These classifications must be regarded with caution given the limited distributional range of each species covered by this investigation. The interspecific mtDNA sequence divergence between O.irroratus and O.angoniensis of 11.57% substaniates morphological, karyotypic and allozymic evidence that these two sympatric species are also sibling species and they appear to have last shared a common ancestor between 1.2 and 2.4 million years ago. / Thesis (M.Sc.)-University of Natal, Durban, 1993.

Pocillopora damicornis is one of the best-studied and most widespread of corals. Previous studies on its sexual reproduction have yielded a diverse and contradictory view of its reproductive strategy. This coral has long been considered a typical hermaphroditic brooder. However, recent studies have demonstrated that its dispersal may also occur through broadcast spawning of gametes. Since both modes of reproduction have advantages for dispersal, P. damicornis reproductive traits are probably driven by environmental pressures. High-latitude habitats are considered marginal for corals since they fall beyond the “normal” limits of coral development. They provide, therefore, a good opportunity to study reproductive strategies that have evolved under different environmental pressures. The pattern and timing of reproduction were studied in P. damicornis on the high-latitude reefs of Durban, KwaZulu-Natal, South Africa (29°S). Branches were randomly selected from colonies on a monthly basis from October 2007 to April 2008. After fixation and decalcification, the coral tissues were processed for histology and cross-sectioned to reveal the presence and stages of gametes within the polyps. The number and size of gametes were estimated using image analysis. P. damicornis in Durban is hermaphroditic and all polyps contained oocytes and spermaries at the end of the breeding season. Since no brooding of embryos or planulae was found in the 175 polyps analysed during the seven-month study period, this coral is presumed to be a broadcast spawner in South African waters. P. damicornis invested seven to eight months in gametogenesis and oogenesis was initiated one month before spermatogenesis. Gametogenesis was confined to the warmer months of the year, commencing during the increase in seawater temperature at the onset of summer. Gamete development was poorly synchronised between colonies during the initial months of breeding, but mature oocyte stages were dominant in all polyps at the end of the breeding season. Spawning was inferred from the disappearance of gametes in April 2008, and may have occurred during full moon in March 2008. This happened before the drop in temperature at the end of summer. P. damicornis in KwaZulu-Natal exhibited a pattern and timing in its reproduction similar to that found in south Western Australia at 32°S. Keywords: coral; Scleractinia; Pocillopora damicornis; high latitude; sexual reproduction; South Africa. / Thesis (M.Sc.)-University of KwaZulu-Natal, 2009.

The ecological role of large herbivores is increasingly recognized due to their ability to influence ecosystem functioning and their impacts on faunal and floral assemblages. Knowledge on the determinants of spatial utilization is crucial towards the successful management of these species according to objectives set for the conservation of biodiversity. While numerous studies have investigated the factors influencing the movements and habitat preferences of large northern hemisphere herbivores, few have focused on members of the African megaherbivore guild. In the context of fenced reserves, elephants have been implicated in the degradation of habitat resulting in negative impacts on biodiversity. Using a kernel analysis approach, I calculated home range size and utilization distributions for five separate herds from an elephant population in Hluhluwe-iMfolozi Park and examined possible determinants. I explored differences in utilization intensity between herds and seasons by comparing the rugosity of utilization distributions. I used a utilization distribution-weighted composition analysis to determine seasonal habitat preferences within the home range and examined the factors influencing daily movement characteristics within different habitats in different seasons. Substantial variation in home range size and location, utilization intensity, habitat preferences and movement responses was evident between herds. Spatial and temporal variation in resource distribution and intra-specific competition explained differences in home range size and utilization intensity. Larger herds underwent more fission-fusion events than small herds, possibly due to resource scarcity and greater competitive interactions. Elephants preferred greener habitats during the dry season and appeared to conform to optimal foraging principles. They utilized forest habitats more than others, selected larger patches with high densities of favoured food items and included greater proportions of common woody species in their diet. Large groups foraged close to rivers in the wet season and appeared to broaden foraging choices in the dry season by moving further away. Restricted displacements in the early morning and evening suggested crepuscular foraging activity while greater displacements at midday could not be explained by the need for water. This study highlights the importance of considering variation in animal movements and habitat utilization in overall conservation planning and when evaluating threats to sensitive habitats, particularly in fenced protected areas. / Thesis (M.Sc.)-University of KwaZulu-Natal, Westville, 2011.

No abstract available. / Thesis (M.Sc.)-University of Natal, Pietermaritzburg, 1987.

Although the leopard (Panthera pardus) is one of the more successful large carnivores, challenges for leopard conservation emulate those of other more endangered carnivores, and they are a model species for investigating issues affecting carnivore persistence worldwide. This thesis represents a six year study of leopard on the Karongwe Game Reserve, South Africa, which provided a unique opportunity to observe various aspects of behavioural ecology in the absence of prey availability constraints or human persecution. Small, enclosed reserves such as Karongwe make up 16.8% of the total land in South Africa and undertaking sound ecological research in these areas provides valuable data for evidence-based conservation and management. The leopard is notoriously shy and difficult to study and I used free darting and habituation to enhance visual observation, in order to understand the ecological processes influencing leopard reproductive success and survival. My results show that leopard in the high prey area studied, consume almost double the number of ungulates as leopard in similar habitats elsewhere. This generalist predator improved its hunting success by selecting vulnerable prey and selectively hunted in habitats of intermediate density, where preferred prey were most abundant. Ample nutrition played a key role in reproductive health and reduced the duration of reproductive parameters below that previously recorded in the literature. Any additional nutritional input could not translate into increased population growth as females were already reproducing optimally. Female territorial size and habitat selection were determined by the availability of riparian habitat and resources of their preferred prey. Territoriality however was governed by prey biomass. Neighbouring leopards were territorial, sharing little space (average 11% territorial overlap) and hunting five times more often in the core than in the rest of their territory. During periods of prey richness, females became more territorial and there was a positive “bottom up” effect through subadult recruitment. Density-dependent intraspecific and interspecific competition for limited space regulated the population around carrying capacity, and constrained population growth. These results provide fundamental baseline data about leopard in the absence of human disturbance, or prey constraints. They highlight that, although the influence of optimal nutrition is important in the reproductive health and territoriality of leopard, habitat quality and quantity are ultimately what govern leopard carrying capacity and population size. I provide baseline reproductive, carrying capacity and territorial data for agencies developing policy, and for setting priorities in conservation and management, as well as habitat protection and restoration, for not only this species but other threatened species as well. / Thesis (Ph.D.)-University of KwaZulu-Natal, Westville, 2013.

This study was conducted to investigate the biology of the Greyheaded Parrot Poicephalus fuscicollis suahelicus in the wild. Field work was conducted in north-east South Africa in the southern limit of the range of the Greyheaded Parrot during two field seasons. Observations from August to December 1999 in the Levubu region, south of the Soutpansberg mountain range, included months of the non-breeding season when Greyheaded Parrots occur seasonally in the area. Observations in the Luvhuvhu-Mutale river confluence area from March to August 2000 included months of the breeding season. No field work was conducted from January to March 2000 due to exceptionally high rains in the southern African sub-region, that prevented access to sites. The Greyheaded Parrot has a widespread distribution, through southern, south Central and East Africa, that has possible changed little in recent years. However, local populations are likely to have suffered extinctions due to habitat destruction and capture for the illegal trade. Populations in protected areas (Makuya Park and Kruger National Park) are less vulnerable to capture than unprotected areas. The conservation of the Greyheaded Parrot outside of protected areas is highlighted. In various parts of the range of the Greyheaded Parrot seasonal movements occur in response to food and nest site availability. Nest sites are possibly limiting in certain parts of its range due to habitat destruction. During post-breeding flocking, the occurrence of larger flocks, possibly family units (mean ± S.E = 4.7 ± 0.2), is common, when birds wander in search for seasonally available food sources. During this period density of Greyheaded Parrots is 0.28 birds/100 ha. Monogamous pairs are more conspicuous during the breeding season (mean ± S.E = 2.1 ± 0.1) and density estimates are 0.14 birds/100 ha. Egg laying is synchronous between pairs with the timed appearance and flocking of juvenile flocks in spring (August/September). Breeding during the dry season reduces competition with other large cavity nesting bird species. Observations suggest that a skewed sex ratio exists in the population (males:females = 2:1). Daily movements are characterised by a bimodal activity pattern. Early morning movements involve flights to activity centres where the accumulation of numerous smaller flocks occur. Here preening, allo-preening and socializing occur with drinking and/or feeding occurring if food and/or water are available. Thereafter, birds move to regular feeding sites, to feed. Activity is decreased during the heat of the day with birds sleeping, resting and/or preening in the canopies of trees. Late afternoon activities involve increased levels of activity and late afternoon return flights to roosts. Greyheaded Parrots are specialist feeders, accessing the kernel of predominantly unripe fruit. This feeding strategy reduces competition with other frugivore competitors. During any one time few tree species are fed on by the Greyheaded Parrot. Greyheaded Parrots were also observed feeding on bark in the breeding season. Two, almost fully fledged, chicks were found dead in a nest and the causes of death undetermined. Their crops contained numerous pieces of masticated bark and insect parts. Behaviours and vocalizations of the Greyheaded Parrot were similar to that recorded in the Cape Parrot. Recognition of the Greyheaded Parrot as a separate species based on species specific calls and DNA warrants further investigation. / Thesis (M.Sc.)-University of Natal, Pietermaritzburg, 2001.

The striped-mouse, Rhabdomys pumilio, is widely distributed throughout southern Africa within a variety of habitats and rainfall regimes. It is found at sea level in the Eastern and Western Cape regions and at altitudes above 2700 m in the Drakensberg mountains. The attraction of R.pumilio to cultivated land and crops has resulted in extensive damage to plants and cultivated crops. A study of the genetic variation between populations of R.pumilio from different regions of southern Africa was undertaken by protein electrophoresis and randomly amplified polymorphic DNA using the polymerase chain reaction (PCR-RAPD). A cytogenetic study was also undertaken. The mean heterozygosity (H=0.074) for R.pumilio was more than twice that estimated for mammals (H=0.036) while the mean percent polymorphism (P=16.1%) was only slightly higher than the mean percent polymorphism obtained for mammals (P=14.7%). The highest heterozygosities were recorded in the Potchefstroom (H=0 .145) and Zimbabwe (H=0 .118) samples and the lowest mean heterozygosity was recorded in the peninsular Western Cape (H=0. 032). A mean Fst value of 0.459 was obtained, suggesting a high degree of genetic differentiation between the samples of R.pumilio but the negative Fis (-0.01) value emphasized that R.pumilio retained an outbreeding population structure. The similarity coefficient between the samples of R.pumilio using PCR-RAPD's ranged between 0.471 and 0.853 and substantiated the argument for genetic divergence between the samples of R.pumilio. An isolation by distance model for the population genetic structure of R.pumilio was supported by the allozymes (r=0.58, p<0.00l) and PCR-RAPD's (0.75, p<0.00l). Temperature and rainfall also had an influence on the allelic frequency distribution of certain loci of R.pumilio. Rogers (1972) genetic similarity varied between 0.796 and 0.988 while the values for Nei's (1978) unbiased genetic distance varied between 0.000 and 0.189 for the different samples of R.pumilio. Subgrouping of the KwaZulu-Natal samples, the peninsular Western Cape and Eastern Cape samples of R.pumilio was evident with the allozymes. With the PCR-RAPD' s the Zimbabwe sample showed the least similarity to the other samples with a KwaZulu-Natal/Potchefstroom subgroup separating from the less well defined Eastern Cape and Western Cape subgroup. Cytogenetic studies of specimens of R.pumilio from some of the localities in southern Africa revealed a chromosomal number of 2n=48 , while the Potchefstroom and Zimbabwe specimens displayed a chromosomal number of 2n=46. Homology in G-and C-banding was recorded. The allozymes, PCR-RAPD's and chromosomal studies suggested subspecies status for the Zimbabwe population of R.pumilio. The Potchefstroom sample displayed a greater genetic similarity to the remaining South African samples of R.pumilio than the Zimbabwe samples and therefore could not be considered for subspecies status. Although the South African samples of R.pumilio displayed a certain degree of genetic divergence, it was insufficient to warrant subspecies status although evolution in this direction was suggested. / Thesis (Ph.D.)-University of Natal, Durban, 1999.