This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. The model was run with variable proportions of the flock untreated (0, 10, 20, 30, 40 and 50%), with ewes selected at random so that reductions in the mean worm burden or egg count were proportional to the treated section of the flock. Treatments to ewes were given either in summer (December; low refugia potential, hence highly selective) or autumn (March; less selective due to a greater refugia potential), and the use of different anthelmintics was simulated to indicate the difference between active ingredients of different efficacy. Each model scenario was run for two environments, specifically a lower rainfall area (more selective) and a higher rainfall area (less selective) within a Mediterranean climatic zone, characterised by hot, dry summers and cool, wet winters. Univariate general linear models with least square difference post-hoc tests were used to examine differences between means of factors. The results confirmed that leaving a proportion of sheep in a flock untreated was effective in delaying the development of anthelmintic resistance, with as low as 10% of a flock untreated sufficient to significantly delay resistance, although this strategy was associated with a small reduction in worm control. Administering anthelmintics in autumn rather than summer was also effective in delaying the development of anthelmintic resistance in the lower rainfall environment where all sheep were treated, although the effect of treatment timing on worm control effectiveness varied between the environments and the proportion of ewes left untreated. The use of anthelmintics with higher efficacy delayed the development of resistance, but the initial worm egg count or number of annual treatments had no effect on either the time to resistance development or worm control effectiveness. In conclusion, the modelling study suggests that leaving a small proportion of ewes untreated, or changing the time of treatment, can delay the onset of anthelmintic resistance in a highly selective environment.