The aims of this study were to verify the efficacy of delayed hormonal treatments performed on day 25 post-insemination on pregnancy rate at 45 and 70 days in buffalo. The trial was performed on 385 buffaloes synchronized by the Ovsynch/TAI protocol and submitted to artificial insemination (AI). Twenty-five days after AI, pregnant animals were assigned to four treatments: (1) GnRH agonist (n = 52), 12 microg of buserelin acetate; (2) hCG (n = 51), 1500 IU of human chorionic gonadotrophin; (3) Progesterone (n = 47), 341 mg of P4 intramuscular (im) every 4 days for three times; (4) Control (n = 54), treatment with physiological saline (0.9% NaCl). Milk samples were collected on days 10, 20 and 25 after AI in all buffaloes to determine progesterone concentration in whey by radioimmunoassay method. Statistical analysis was performed by anova. Pregnancy rate on day 25 after AI was 52.9%, but declined to 41.8% by day 45, indicating an embryonic mortality (EM) of 21%. If only control group is considered, the incidence of EM was 38.9%. Pregnant buffaloes had higher (p < 0.01) progesterone concentrations on day 20 and 25 after AI than both non-pregnant buffaloes and buffaloes that showed EM. The treatments on day 25 increased (p < 0.01) pregnancy rate, although in buffaloes with a low whey progesterone concentration on day 20 and 25 after AI (n = 22); all treatments were ineffective to reduce EM.