It has been established that prostaglandin (PG) F-2\g=a\induces the involution of corpora lutea (CL) in hysterectomized guinea-pigs (Blatchley & Donovan, 1969) and that the guinea-pig uterus is probably the major source of this luteolytic factor (Poyser, 1972). However, the measurement of PGF-2\g=a\levels in the utero-ovarian venous plasma of cyclic guinea-pigs (Blatchley, Donovan, Horton & Poyser, 1972) revealed that, even though the concentration was low, PGF-2\g=a\was present throughout the oestrous cycle and that there was asynchrony in luteal regression (Days 12\p=n-\16: Rowlands, 1956) and the peak of PGF-2\g=a\secretion (Day 15: Blatchley et al., 1972). During experiments to study the effects ofPGF-2\g=a\on the different tissue components of the ovary, it was found that the CL responded to the luteolytic action of PGF-2\g=a\only at certain times.Guinea-pigs were hysterectomized before puberty (24-32 days old) and then injected i.p. twice every day with 0-125 mg PGF-2a (equivalent to 0-168 mg THAM salt, Upjohn Co.) in 0-5 ml 0-9% NaCl, i.e. 0-057-0-100 mg PGF-2a/100 g body weight/day (Group A). Other guinea-pigs of the same age were (i) hysterectomized and injected twice daily with 0-5 ml NaCl (Group B), (ii) subjected to sham-operation, i.e. sectioning of the utero-vaginal junction but leaving the uterine horns in situ, and similarly treated with NaCl (Group C), and (iii) given no treatment (Group D). Most animals were killed in the second oestrous cycle, but 3 Group-and 3 Group-D animals were maintained until the early stages of the fourth cycle. At autopsy the ovaries were fixed and serially sectioned for histological observation. The CL volume was calculated from the mean of 3 diameters set at right angles to one another, measured with an ocular micrometer from the ovarian sections. Oestrous cycle length was calculated by daily examination of the vaginal closure membrane and the day after the first day the vaginal membrane was fully perforate was taken to be Day 1 of the cycle.The first vaginal opening was observed in animals of Groups C and D when they were 36-2 ± 0-8 (N = 23) days old (mean ± S.E.M., the total of 11 animals in Group C and 12 animals in Group D), but the first vaginal opening occurred at 45-2 ± 2-9 (N = 14) days of age in the PGF-2a-treated animals in Group A. It is not clear whether this delay was due to the combined effect of hysterectomy and PGF-2a treatment or just the shock of hysterectomy as the amount of blood loss in hysterectomy was always greater than in the sham-operation. However, once the first oestrus had occurred the cycle lengths in Group A were, for the first to third cycle, 14-9 ± 0-6 (number of cycles observed = 12), 15-0 ±1-2 (3) and 16-3 ± 0-9 (3) days, respectively; and these durations were not significantly differ¬ ent (Student's / test) from those of the control animals in Group D (15-4 ± 0-6 (9), 16-3 ± 0-3 (3) and 15-7 ± 0-3 (3) days, respectively). There was also no significant difference between the duration of the first cycle in Groups C and D. Continual t...