Mistletoes are a group of hemiparasitic plants that grow on a wide variety of host trees and differ in their degree of host specificity, ranging from specialists to generalists. Mistletoes can also be locally host specific where host preference varies geographically, i.e. at a given location a mistletoe species may infect only part of its overall host set. The mistletoe Agelanthus natalitius parasitises at least 11 tree genera distributed throughout South Africa. However, there is geographic variation in infection patterns over the parasite’s range, suggesting that A. natalitius may be locally host specific. We quantified the degree of host specificity and tested the mechanisms that direct host specialisation in two distinct mistletoe populations at Highover and Mtontwane (about 110 km apart) in KwaZulu-Natal, South Africa. We investigated the distribution, abundance and community composition of woody species that host the mistletoe. We also assessed the effect of light on germination and early survival of the mistletoes in a greenhouse experiment. We conducted field reciprocal transplant experiments at both sites to investigate the compatibility of these mistletoes with their hosts Acacia karroo and A. caffra during early development. We then analysed the nutrient and water contents of the mistletoe-host pairs to investigate the role of nutrient and water status in directing host specificity in mistletoes. We further studied avian dispersal in the field and in captivity to investigate optimal dispersal distance and germination success, and evaluated their role in determining mistletoe host specificity. At both study sites, five host species were recorded as being parasitised by the mistletoe A. natalitius. A. karroo and A. caffra appear to be the two most common host species in the region; both grow abundantly at the study sites and were recorded with high infection by A. natalitius. However, A. karroo is the most abundant host species and the mistletoe showed a high degree of host specificity on A. karroo. Infection by mistletoes was positively correlated with tree size, and was highly aggregated, both individually and locally. Field observations and greenhouse shade experiments showed that light can influence mistletoe distribution. Germination of mistletoe seeds was independent of host species and site. However, hypocotyls (the structures that develop into haustoria) grew longer when placed on their source host species within their locality. Additionally, they showed preference for the most abundant host species, A. karroo. Water and nutrient status of the host species A. karroo and A. caffra had no significant effect. Thus, host nutrient and water content may not account for host specificity in this mistletoe species. Mistletoes accumulated more nutrients and maintained more negative than their host trees. We also investigated the mistletoes’ use of passive nutrient uptake (from host xylem) and active nutrient uptake (from host phloem) by using the N:Ca ratio as an index of nutrient access. Mistletoes growing on A. caffra had a ratio > 1, i.e. the mistletoe actively accessed nutrients from the phloem of host trees. However, mistletoes on A. karroo had a N:Ca ratio < 1, which implies that they passively accessed nutrients from the xylem. The difference in mechanism of nutrient acquisition on different host species may reflect the level of compatibility between mistletoe and host. Several bird species were frequently observed to feed on mistletoes, many of which were used in our captivity studies. Although birds did not consume mistletoe fruits in captivity as they do in the field, they were effective in removing the pulp cover of mistletoe fruits and exposing seeds in germinable condition. In captivity, the Red-winged Starling ingested whole fruits and regurgitated seeds, deliberately wiping their bills on twigs to remove the sticky seeds. As a result, germination success of mistletoes processed by Red-winged Starlings was higher than any other bird species tested in captivity. Overall, there appears to be host specificity in morphologically identical mistletoes. Understanding the mechanisms that result in host race evolution are potentially important to the process of speciation in hemiparasitic mistletoes. We need to take into account genotypic matching in conserving these different forms of mistletoes and their host Acacia genotypes. Further research into the mechanisms of host specificity and patterns of genotypic matching is warranted. / Thesis (M.Sc.)-University of KwaZulu-Natal, Pietermaritzburg, 2009.