Summary. Rams and ewes of the Romney Marsh (N = 6), Dorset Horn (N = 8) and Australian Merino (N = 8) breeds were subjected to 4 successive periods of alternating 6 h light/18 h dark ('short' days) and 18 h light/6 h dark ('long' days) preceded by 16 weeks of 12 h light/12 h dark. The initial period was of 32 weeks (16 weeks 'short' days; 16 weeks 'long' days) and the next 3 were of 24 weeks (12 weeks 'short' days; 12 weeks 'long' days).
Rams of all breeds showed a cyclic pattern of growth and regression of testes associated with plasma testosterone concentration, influenced by the change in light regimen 15–19 weeks previously. Sexual behaviour was also cyclic but lagged by some 6–7 weeks. The changes were greatest in the Romneys and least in the Merinos in which a higher degree of sexual activity was evident even when the testes were regressed (P < 0·001). This was the major breed difference.
All ewes of the Romney and Dorset breeds showed marked seasonality related to the imposed light regimen, whereas only 1 of the 4 Merinos did so. The mean peak of ovarian activity in the former 2 breeds coincided with that of maximum sexual activity of rams housed with them; that is, some 6 weeks after maximum scrotal volume.
The rams and ewes were subjected to serial blood sampling episodes for plasma LH and testosterone and tested for plasma LH release following GnRH administration. There was little variation between breeds in LH concentration. Testosterone concentration varied greatly in the ram, highest levels associated with the developed phase of the testes and with maximum LH pulse frequency. The LH response to GnRH changed with respect to the state of the gonads. Maximal responses were observed in the developing phase of testicular growth although this variation was greater in the Romney and Dorsets than in the Merinos (P < 0·001). In the ewes, maximal responses were seen in the follicular phase (P < 0·001), with no difference between the luteal and acyclic phases. There were no breed differences.
Plasma pooled from the serial blood sampling episodes was assayed for prolactin. Rams of each breed exhibited a cyclic pattern of plasma prolactin concentration, directly related to photoperiod (P < 0·001) and inversely related to testicular activity (P < 0·001). All Romney and Dorset ewes, but only the 1 'seasonal' Merino ewe, displayed a cyclic relationship between plasma prolactin and photoperiod (P < 0·001). Mean prolactin concentrations were greater in the Romney and Dorset rams (P < 0·02) and ewes (P < 0·05) than in the Merinos.
It is concluded that, although these 3 breeds followed a similar endocrinological pattern, there was a marked difference in the 'ultimate' responses (i.e. libido and oestrus) with the Merino displaying the least constraint to the imposed light regimen. This could reflect the natural and artificial selection pressures upon the Merino for unrestricted (year-round) breeding.