Sixteen rams were used to quantify the effects of long days, imposed during late winter/early spring, with or without exogenous melatonin, on plasma testosterone concentrations and ram serving capacity. Rams were assigned to two groups: photoperiod-treated rams (Artificial Photoperiod, AP; n = 8), exposed to 2 months of long days (16 hr of light/day) between 22 December and 22 February, and control rams (Natural Photoperiod, NP; n = 8). At the end of the long-day period, AP rams were returned to the natural photoperiod, and each ram in the two groups either did (+M) or did not (-M) receive three subcutaneous melatonin implants. Four groups were created as follows: AP+M (n = 4), AP-M (n = 4), NP+M (n = 4) and NP-M (n = 4). Thirty days after of the onset of photoperiodic treatment, AP rams (13.5 ± 2.8 ng/ml) had significantly (p < .05) lower testosterone levels than NP rams (36.7 ± 1.0), and similar differences were not apparent at the end of the photoperiod treatment. A month later, AP rams (24.3 ± 7.9) had higher (p < .10) testosterone levels than NP rams (13.1 ± 5.0), with no effect of melatonin treatment. Fifty days after melatonin implantations, rams were exposed for 20 min to three oestrous ewes. AP rams (2.50 ± 0.42) exhibited significantly (p < .05) more serves than did NP rams (1.11 ± 0.39), and melatonin treatment had no significant effect; however, the interaction between treatments was significant. Time to first serve was significantly (p < .05) shorter in AP (2.30 ± 1.20 min) than it was in NP rams (5.58 ± 0.68 min). In conclusion, exposure to 2 months of long days in late winter/early spring, with a return to natural day length of shorter duration, increased plasma testosterone concentrations and sexual performance in rams with or without exogenous melatonin. This particular management is an option if a non-hormonal reproductive strategy is scheduled; yet, if the use of exogenous hormones is feasible, melatonin implants increase the mating efficiency of rams.