The aim of this study was to evaluate the possible relationship of pharmacological induction of estrous and/or ovulation with the occurrence of twin pregnancies in Thoroughbred mares. Out of 680 mares, 356 received one of the following treatments during the estrous cycle in which they became pregnant: injection of 0.5mg of cloprostenol at the ultrasonographic detection of a CL (n=86); injection of 5000 IU human chorionic gonadotropin (hCG) immediately before mating (n=221); injection of 0.5mg of cloprostenol at the ultrasonographic detection of a CL plus injection of 5000 IU hCG immediately before mating on cloprostenol-induced estrous (n=49). The other 324 mares, not treated for induction of estrous or ovulation in the estrous cycle resulting in pregnancy, were used as control group. The occurrence of twin and single pregnancies in treated and control mares underlines that the percentage of twin pregnancy in treated mares (16.6%) was statistically significantly higher (P<0.0001; odds ratio, OR=2.87) than the percentage of twinning in the control group (6.5%). Comparison of the occurrence of twins between treatments revealed a statistically significant difference between mares treated with hCG alone compared to animals given prostaglandin F2alpha (PGF2alpha) plus hCG. The results show a statistically significant difference for each treatment compared to controls, with the least difference (P<0.05; OR=2.18) for the comparison between hCG treatment group and controls, a significance of P<0.01; OR=3.05 for the comparison between PGF2alpha treatment and controls, and a highly statistically significant difference (P<0.0001; OR=6.37) for the comparison between PGF2alpha plus hCG-treated animals and controls.