Embryo quality is influenced by culture conditions, which affect IVM, IVF, and IVC rates. One of the most efficient ways to evaluate the embryonic quality of IVP blastocysts is by differential staining of inner cell mass (ICM) and trophoblast (TF). Bovine embryos of superior quality should present the total number of cells close to the number of cell cycles (Neuber et al. 2002 Theriogenology 57, 2193-2202). In this study, we analyzed the effects of fetal bovine serum (F) and bovine serum albumin (B) replacement for ovalbumin (O) on nuclear maturation, cortical granule migration, pronuclear development, blastocyst rates, and differential staining of ICM and TF in Day 7 blastocysts. The treatment groups were named as follows: the first letter is the protein source used for IVM, the second for IVF, and the third for IVC. When 2 protein sources were used in the same step, the plus symbol (+) was used. The oocytes were IVM in TCM-199, supplemented with the following: 10% F, or 4 mg mL-1 B, or 4 mg mL-1 O, and 1.0 Âμg mL-1 of FSH, 50 Âμg mL-1 of hCG, 1.0 Âμg mL-1 of estradiol, 0.2 mM sodium pyruvate, and 83.4 Âμg mL-1 of amikacin. IVF was accomplished in TALP-IVF medium, with 0.2 mM pyruvate, 83.4 Âμg mL-1 of amikacin, and 6 mg mL-1 B or O. IVC was in SOF, with F, B, or O. The control group (CONT) consisted of the treatment FBF + B. Pronuclear development was compared by the chi-square test, whereas the other results were analyzed by ANOVA followed by the Tukey test, using SAS at 5% significance level (SAS Institute Inc., Cary, NC, USA). For IVM, the treatments F, B, O, and B + O did not affect (P > 0.05) nuclear maturation (73.92 to 78.78%) and cortical granule migration rates (58.89 to 66.76%). Regarding pronuclear development, the treatment FO (76.67%) was similar (P > 0.05) to the control group (82.95%), which was superior (P < 0.05) to the treatments BB (56.98%), BO (39.02%), OB (37.36%), and OO (39.24%). Blastocyst rates in FBF (42.8%) and control (45.0%) groups were superior (P < 0.05) to treatment OOO (26.0%) but similar (P > 0.05) to FOF, BBB, BOB, and OBO (32.0 to 35.8%). The average of blastocyst ICM cells of the group OOO (16.79) was inferior (P < 0.05) to the other groups. However, the average of TF cells on blastocysts of the group OOO (38.25) was similar (P > 0.05) to the groups BBB (45.74) and BOB (45.60) and inferior (P < 0.05) to the groups CONT (57.59), FBF (54.41), FOF (56.74), and OBO (47.35). The total average cells in the blastocysts of the group OOO (56.04) was inferior (P < 0.05) to the groups CONT (84.86), FBF (78.96), FOF (81.32), BBB (68.11), BOB (69.55), and OBO (69.82). The total cell number in the treatments, with several sources of protein supplementation, varied from 56.04 to 84.86. Considering the evaluation interval, this average cell number was discreetly inferior to that expected for the chronological age of the blastocysts. We concluded that it is possible to produce bovine embryos in the absence of F and/or B, with the protein source O, although it reduced blastocyst rates when used in all 3 steps of embryo in vitro production and resulted in blastocysts of inferior quality. Financial support: FAPESP 05/60389-2 and CNPq.