To evaluate the effect of gastrointestinal parasites on grazing behaviour, herbage intake and milk production in spring calving dairy cows, 12 naturally infected control cows were compared with 12 similar animals treated on three occasions (June, July and September) with eprinomectin. The cows were blocked according to calving date, parity, live weight and milk yield during week 2 after turnout and then allocated to the treatments. The grazing area was sub-divided into two sets of 12 replicated paddocks of equivalent size and topography. Pairs of either control or treated animals were randomly assigned to graze each paddock over the duration of the study. Within each plot, the pair of cows grazed a series of 1-day paddocks, of areas calculated to provide 72 kg of herbage dry matter measured to ground level. Faecal samples were collected from each cow in April, prior to allocation, and every 28 days thereafter. Samples were submitted for counts of nematode eggs (sensitivity 1 epg) and the presence of Dictyocaulus viviparus larvae. Additional faecal samples were taken on each occasion for culture and nematode identification. Pasture samples for direct larval counts were collected at the same time as faecal sampling. Behaviour measurements on all cows were made during three periods, once before the first treatment with eprinomectin and thence after the 2nd and 3rd treatments. During each behaviour measurement period, grazing and ruminating behaviour were recorded over two 24-h periods and measurements of components of short-term intake rate were made during a morning and a late afternoon grazing meal. Milk yield was recorded daily and milk quality was recorded weekly. Live weight and body condition score were recorded on the day of allocation, the day of initial treatment and thereafter at weekly intervals until the end of the trial. The parasitological results showed low levels of faecal egg output throughout the study with group arithmetic means ranging from 0 to 6.8 epg. Faecal culture yielded predominantly larvae of the genus Ostertagia, but the following genera were also identified: Cooperia, Oesophagostomum and Trichostrongylus. Pasture larval levels were also low with peak values of 135 and 58 L3/kg DM herbage (7 August) in the paddocks grazed by the control and treated cattle, respectively. Thereafter, larval counts on paddocks grazed by treated cows declined to undetectable levels by October, while control paddocks remained at approximately 40 L3/kg DM. There was no effect of treatment on components of grazing or ruminating behaviour recorded over 24 h or on short-term intake rates. There were significant differences between components of short-term intake rates measured during the morning and afternoon grazing meals. The overall milk yield response to treatment with eprinomectin was +1.68 kg/day solids-corrected milk (SCM) (P=0.026). The overall response included significant (P<0.050) increases in mean daily SCM yield following each of the three treatments, indicating a positive response to repeated treatments at several different stages of lactation. There were no significant differences in the overall percentages of fat, protein or lactose between control and treated groups. The differences in live weight were not significant, although there was a consistent pattern throughout for the treated cows to be heavier than the controls.