The effects of melatonin (implants, M or no implants, C) and plane of nutrition (high, H or low, L) on mammary development and growth hormone (GH) concentrations were investigated in prepubertal Boutsiko mountain breed ewe lambs. Eighty female lambs were assigned to each of 4 treatments: ad libitum feeding control (HC), HM, LC and LM. The rearing treatments started and ended at mean ages of 63 and 160 d, respectively. Feed restriction resulted in a mean daily gain of 70.6% of the ad libitum-fed lambs during the experimental period. Melatonin (18 mg Regulin) was administered at 68 d of age (January 10) and replaced on March 1. Blood samples were collected from 10 lambs in each treatment group at the end of the experiment for GH measurements. At a mean age of 160 d, seven lambs from each treatment group were slaughtered and the udder was removed. One udder half was trimmed and the parenchyma and fat pad portions were kept for determination of deoxyribonucleic acid (DNA) content. Melatonin did not influence mammary development parameters, while the mass of parenchyma tended to be greater in lambs on low than high nutrition planes (P<0.10). Mean mammary parenchymal weight and DNA content were 25.1 and 29.2 g and 52.5 and 58.2 mg in high and low nutrition lambs, respectively. Mean plasma GH concentrations were not affected by melatonin treatment and were higher in low than high nutrition lambs (P<0.01). There were no correlations between mean plasma GH concentrations and parenchymal DNA content, or between mean daily weight gain and parenchyma (g), in contrast to those found in a previous experiment with lambs of the same breed but greater age at slaughter. The results suggest that a period of accelerated mammary development occurs later than 140 d of age in Boutsiko mountain breed ewe lambs.