My interest in the behaviour and movements of birds of arid and semi-arid ecosystems began when my wife, Sue Milton, and I were Roy Siegfried, Director, at that time, of the Percy approached by Prof. FitzPatrick Institute of African Ornithology, to set up a project to investigate granivory in the South African Karoo. Sue and I spent some time finding a suitable study site, setting up accommodations and an automatic weather station at Tierberg, in the southern Karoo near the village of Prince Albert, and planning projects. Among our first projects was a transect where we noted plant phe nology, measured seed densities on the soil surface, counted birds, observed ant activity, measured soil surface temperatures and col lected whatever climate data we could at 40 sites along a 200-km oval route. Along the way, we became interested in the marked presence and absence of birds at certain sites - abundant birds one day, and very few birds at the same site a month later. Subsequent counts along fixed transects through shrublands confirmed that a number of bird species were highly nomadic over short and long distances, locally and regionally, leading to speculation on how widespread these movements were in the arid ecosystems of the world.