Tourism is the most recent large-scale human activity in the Antarctic. The 1991 Protocol on Environmental Protection to the Antarctic Treaty requires that all activities in the Antarctic, including tourism, shall "be planned on the basis of information sufficient to allow prior assessments of, and informed judgements about their possible impacts ... and ... regular and effective monitoring shall take place to allow assessments of the impacts of ongoing activities." As yet there is an acknowledged lack of hard data on the effects of tourism on the Antarctic environment, and no such monitoring programme exists. Because of its scale and environmental context, shipborne tourism is likely to disturb Antarctic ecosystems. 96.5% of all Antarctic tourists are shipborne and over 90% of their visits are to the Maritime Antarctic (Antarctic Peninsula and South Orkney and South Shetland Islands), Antarctica's ecologically richest area. This study is founded on the author's five years of research, both at a field station and as a shipboard expedition leader. The programme formed part of a longer study of polar tourism, Project Antarctic Conservation, directed by Dr Bernard Stonehouse of the Scott Polar Research Institute, University of Cambridge. Assessing three phases of shipboard operation - on the ship, in landing craft, and ashore - operations ashore were found to be the most difficult to quantify and likely to show cumulative effects from repeated small perturbations. Thus, landing operations and sites became the main focus of investigation. Three major research objectives were: (1) to examine patterns in landing site use, based on NSF/IAATO data; (2) to assess the industry' s landing site organisation and site selection procedures based on field experience; and (3) to investigate ecological disturbance from tourist visits at a popular landing site: for this Cuverville Island (64°4l'S, 62°38'W) was selected and studied for three consecutive seasons 1992-95. A total of 128 landing sites were identified, clustered into five geographic areas, and found to differ widely in physical characteristics, level of use and ecological vulnerability. Trends in numbers of passenger, voyages, sites visited etc., over the past decade, were examined. Each voyage involves visits to several sites, to show tourists a range of environments and wildlife. Small vessels, carrying fewer passengers on shorter voyages, have recently come into vogue, each vessel making more voyages per season. Yet, as the number of landings per voyage has not decreased, a recent slight decrease in the annual number of passengers, has not resulted in a proportionate decrease in the number of landings per season. Landing operations are generally competent, safe and environmentally sensitive. However, there is no generally-accepted measure of site vulnerability, and no formal recognition that sites vary in sensitivity; the industry's own recommendations for site management are uniform for all sites. Expedition leaders, who ultimately decide which sites are visited, vary widely in awareness of differences in site vulnerability, but are left to develop their own visitor management strategies at individual sites, based on personal perceptions. Thus, management of landing sites, insofar as it exists at all, is informal, and undertaken by individuals within the industry. Ecological studies of Cuverville Island are described and analysed. Visits were on the whole well managed, and team research under the author's field leadership revealed few measurable adverse effects on local breeding populations of wildlife. However, possibilities exist for long-term impacts arising from repeated use. From these studies have been developed key management parameters for assessing visitor impact, and designing effective measures for long-term monitoring of vulnerable sites. The author concludes that, for effective long-term management of sites, management objectives must be defined, and variations in sensitivity and vulnerability recognised. Sites that appear to be at risk from shipborne tourism must be assessed individually. The dissertation proposes site monitoring objectives, practical methods for site assessment, and a framework for long-term monitoring and management programmes, that are both consistent with the requirements of the Protocol, and appropriate for practical implementation.