Lactating dairy cows (n = 72) with a mature corpus luteum (CL) (diameter of > or = 17 mm) determined by ultrasonography and having a follicle with a diameter of > or = 10 mm were randomly assigned to four groups. Cows were treated with cloprostenol i.m. once or twice, or with dinoprost i.m. once or twice with an 8-h interval. The ovaries of each cow were scanned daily by transrectal ultrasonography to measure the changes in the areas of CL and the largest follicle and to determine the occurrence of ovulation. Oestrus was verified twice daily. In addition, blood sample was withdrawn from each cow daily for measuring progesterone (P4) concentrations. Significant decreases in the percentage changes relative to areas of CL and P4 concentrations or increases in the percentage changes in the area of the largest follicle on day 0 were detected in each group during the experiment. However, the type of the drug and the number of the treatments had no significant effect on those parameters. Cows ovulated with or without showing oestrus (group A) and cows exhibiting no oestrus and ovulation (group B) were also evaluated. In contrast to the mean area of the CL and the mean concentration of P4 on day 0, the mean area of the largest follicles between the two groups on day 0 differed significantly. Significant decreases in the percentage changes relative to the area of the CL and P4 concentration or increases in the percentage changes relative to the area of the largest follicle during the experiment were detected in both groups; however, there were no group differences. Treatment of dairy cows with two injections of prostaglandins (cloprostenol or dinoprost) at an 8-h interval resulted in more cows being observed in oestrus within 5 days after treatment and having significantly higher pregnancy rate than those treated with a single prostaglandin injection.