Summary The precision of embryo evaluation using stereoscopic microscopy (SM) and inverted phase contrast microscopy (PCM) was compared in 20 Bos indicus cows superovulated at two different times of the year. In total, 118 embryos were collected and classified according to their developmental stage and quality by two independent evaluators using SM and inverted PCM. Cohen's kappa coefficient was used to determine concordance between SM and PCM observations. A good level of agreement (k = 0.616) was found for quality level, and a moderate one (k = 0.464) for developmental stage, particularly at the morula stage. Using the TUNEL (terminal deoxynucleotidyl transferase dUTP nick-end labelling) technique, concordance level was deemed to be low with the SM (k = 0.169), and poor with the PCM (k = 0.217). Differences in concordance levels were also found between observations made at the two times of year, 78 embryos were evaluated in the rainy season when concordance level was good (k = 0.68), in contrast to the 40 embryos evaluated in the dry season when agreement was found to be poor (k = 0.24). In conclusion, inverted PCM was somewhat more effective for evaluating embryos, particularly at the morula stage. However, considering the high cost of an inverted PCM, the differences observed do not justify its purchase for routine embryo evaluation.