A computer simulation model of ovine fascioliasis was developed. The simulated flock, comprising a maximum of 60 non-reproductive sheep, grazed on a perennial pasture with a suitable habitat for the intermediate host snail Lymnaea tomentosa at the low end of a ‘gradient’ across the simulated 2-ha area. The driving force for both herbage production and the life cycle of Fasciola hepatica is meteorological information—comprising maximum and minimum daily temperatures and moisture (rainfall and any water supplied by irrigation). The model simulates intake of both herbage and metacercariae, with sheep seeking green herbage, if available, in preference to dry material. This automatically produces seasonal variation in the grazing pattern and hence in access to metacercariae. The production of wool and liveweight gain in sheep are simulated and include the interaction with fluke burden. Flock replacement is also dealt with and is made to closely replicate commercial management practices. The life cycle of F. hepatica is simulated in detail and the role of the intermediate host is dealt with by permitting the snail production to be in one of three states of physical activity, only one of which is suitable for completion of the development and release of cercariae. From information arising from the simulation, an economic comparison of the control strategies under test was made and the results printed. The completed model was validated against the results of a field study conducted by the authors at Prairie, Victoria, Australia (Meek & Morris, 1979) using meteorological data which had been measured at the field site as the model's driving force and then comparing the predictions of the model with the actual field measurements. The model was also validated against the results of a field investigation conducted in New South Wales, Australia, by Boray et al. (1969). The degree to which the model conformed with expectations and with the actual field data was assessed. In two respects the agreement between model predictions and field data was inadequate. First, the model overestimated the weight gain of the simular sheep on irrigated pasture under conditions of severe drought; secondly, simular snails were unexpectedly sensitive to changes in soil moisture. These aspects require further examination. Sensitivity analyses were also performed on the model. It was determined that simular financial return was sensitive to both stocking rate and the proportion of the simular area defined as snail habitat. Therefore both these factors warrant consideration in the choice of control strategies. Various control measures were simulated, by means of the validated model, either singly or in combination, and were assessed by means of the expected annual margin over variable husbandry costs obtained over the course of a five-year simulation period, from a standard flock of 100 sheep. With respect to the phophylactic use of anthelmintics, the model provided a method of ranking various alternatives at each of a number of combinations of stocking rate and snail habitat size. Under almost all simular circumstances tested, five to six strategically timed treatments produced the greatest economic benefit from controlling the disease. The model predicted that it was possible to eradicate ovine fascioliasis from a closed system if the recommended treatment strategy were rigidly followed over a number of years. Simular results also suggest that molluscicides may be integrated to economic advantage with anthelmintics in instances where the grazier opts for an anthelmintic treatment strategy with fever than the recommended optimal number of treatments per annum. It is also suggested that integration of rotational grazing with the use of anthelmintics warrants further investigation to ascertain if its apparent benefits can be realised in practice. It was concluded that the computer simulation model provides a formal framework within which alternative control strategies can be assessed and that it provides a basis for informed decision-making about potential control methods. Its predictions can now be tested under practical conditions.