Abstract: Maternal undernutrition during gestation can condition offspring adult health, with the periconceptional period pointed out as a key period. The aim of this study was to evaluate the effects of maternal periconceptional undernutrition on pregnancy and offspring growth performance in sheep. 52 Merinos d'Arles ewes were fed to requirements (control group, C), whereas 64 ewes received 50% of their dietary needs from -15 to +30 days post-conception (restricted group, R). Thereafter, both groups were fed according to needs. Maternal body weight (BW), body condition score (BCS) and Non Esterified Fatty Acids (NEFA), progesterone, leptin and cortisol plasma concentrations were monitored weekly during the restriction period and the following month, then monthly until weaning. Lambs were weighed weekly until weaning at 22 kg BW, then monthly. Plasma leptin was monitored monthly in lambs. The BW, BCS, and leptin concentrations were significantly decreased, whereas NEFA and cortisol concentrations were increased in R dams. Maximum progesterone concentration was higher in R ewes that had a high (10-25%) vs. low (0-10%) BW loss during restriction (27.9 +/- 2.59 vs. 20.8 +/- 2.00 ng/mL, P < 0.05). Overall, gestation was significantly longer in the R group (151.0 +/- 0.3 vs. 149.4 +/- 0.4 days, P < 0.001). There was no difference between groups for pregnancy rates, prolificacy, birth weight and lamb mortality, but the proportion of male lambs was significantly higher in the R group, only for singletons (16/26 vs. 9/26, P < 0.05). Lamb growth was not significantly modified by treatment. Leptin concentrations at birth were significantly lower in R vs. C males (6.15 +/- 0.13 ng/mL vs. 7.42 +/- 0.36 ng/mL, P < 0.05), whereas in females, leptin concentrations were significantly higher in R vs. C lambs at 4 mo of age (7.31 +/- 0.27 ng/mL vs. 6.41 +/- 0.29 ng/mL, P < 0.05). These results indicate that maternal periconceptional undernutrition in a hardy breed does not significantly affect lamb birth weight and growth rates, in contrast to previous reports in other breeds, suggesting that caution must be taken when extrapolating programming data between breeds and breeding conditions. (C) 2012 Elsevier Inc. All rights reserved.