The most important barrier to the increase of buffalo productivity is an overall poor reproductive efficiency, characterized by late sexual maturity, seasonal anestrus and long periods of postpartum ovarian inactivity resulting in extended calving intervals and poor expression of estrus behavior (Singh J et al., 2000, Anim. Reprod. Sci. 60–61, 593–604). Buffaloes are seasonal breeders with the highest reproductive activities during winter (short day lengths) and a high frequency of anestrus during the summer months (Singh G et al. 1985, Ind. J. Anim. Res. 19, 57–60). Recent research demonstrated that a combination of progesterone, estradiol benzoate and equine chorionic gonadotropin (eCG) was effective for estrus induction and synchronization in buffalo heifers under Mediterranean conditions (Barile et al. 2001, Livestock Prod. Sci. 68, 283–287). The aim of the present study was to investigate the impact of an estrus synchronization protocol on reproductive efficiency of water buffalo during out of the normal breeding season. A total of six heifers (21 to 23 months of age) and three cows (5, 6 and 18 years of age) were enrolled in an estrus synchronization protocol lasting for 12 days. All animals were kept under tropical conditions in the coastal part of Paraná (Antonina), about 450 km south of São Paulo. The experiment was performed in December, 2002, during the Brazilian summer season, when reproductive efficiency of buffaloes is greatly reduced. On the first day of the protocol (Day 0), animals were implanted with an intravaginal device containing 1 g of progesterone (DIB, Syntex SA, Buenos Aires, Argentina) and injected with 10 mg estradiol benzoate (Estrogin, Famavet, São Paulo, Brazil). On Day 9, the DIB implant was removed and the animals received 150 μg (i.m.) of cloprostenol (Prolise, Syntex SA, Buenos Aires, Argentina) and 2500 IU of eCG (Novormon, Syntex SA, Buenos Aires, Argentina). On Day 11, all animals received 1500 IU of hCG (Vetecor, Lab. Calier, Spain). Artificial insemination (AI) was performed on Day 12 using frozen-thawed semen from a bull of proven fertility. Only one AI was performed per heifer/cow. Pregnancies were determined by ultrasound examination at 53 days following AI and confirmed by rectal palpation at 90 days post AI. The use of this estrus synchronization protocol, followed by fixed-time insemination, resulted in four pregnant heifers (66%) and three pregnant cows (100%). Our results demonstrate that buffalo reproduction can be successful during out-of-breeding season when adequate hormonal treatment is used. Additional experiments should be done to validate the protocol.