The physical and financial ramifications of using real-time ultrasonic scanning for the differential management of single- and twin-bearing ewes have been evaluated. A model of a self-replacing flock of Merino ewes located in Western Victoria was used. A number of different management systems involving scanning were tested under a variety of seasonal and market conditions. Local weather, farm survey and research data were used in the analysis. Scanning enables feed to be used more efficiently. Thus, economic advantages to scanning were evident where the feed supply was limiting during pregnancy and lactation with the greatest financial response recorded during drought. Ewes in southern Australia which lamb in the autumn regularly experience extended periods of feed shortage and are therefore particularly sensitive to the management of the available feed. Returns to scanning ewes lambing in May were therefore generally higher than for ewes lambing in August when pasture availability was less likely to be limiting. The identification of twin-bearing ewes became an economically more viable proposition as their proportion in the flock was increased. Increases in farm income were also achieved when scanning was used to improve flock fecundity levels. Setting the fecundity of a flock at a very high level resulted in a marked increase in farm cash operating surplus. The susceptibility of highly fecund flocks to erosion of their reproductive potential, and the financial penalties of not preventing such wastage through technologies such as scanning, are demonstrated. For flocks where external sources of higher fecundity are unavailable, or in some way undesirable, scanning provides a practicable and financially viable means of increasing the size of the breeding flock. However, given the relatively low levels of fecundity in Australian wool-producing flocks, the increase in net farm income resulting from scanning without substantial changes in genotype and management would in general be small. Farm income was more sensitive to a change in wool production than it was to an equivalent change in the number of stock sold in a given class. Even where a dramatic fall in the price of wool was maintained for a prolonged period, the improvement in farm income resulting from scanning a flock with few twins was modest. The sale of non-pregnant ewes after scanning in some or all years was not economic. Such ewes produced up to 20% of the flock's wool. The price received for non-pregnant ewes would have to be much higher than was assumed in our study before this strategy would become viable. A major aim of this study was to evaluate scanning technology and identify those areas where it is economically viable. This aim was achieved, but another important insight emerged. For the benefits of scanning to be maximised, the opportunities it provides to increase selection pressure for improved fecundity, reduced reproductive wastage and increased wool production must be utilized.