Day 7 bovine somatic nuclear transfer (NT) embryos reconstructed from granulosa cells were examined for numerical chromosome aberrations as a potential cause of the high embryonic and fetal loss observed in such embryos after transfer. The NT embryos were reconstructed using a zona-free manipulation method: half-cytoplasts were made from zona-free oocytes by bisection, after which two half-oocytes and one granulosa cell (serum-starved primary culture) were fused together and activated. The NT embryos were cultured in modified synthetic oviductal fluid containing essential and nonessential amino acids, myoinositol, sodium citrate, and 5% cattle serum in microwells for 7 days, at which time nuclei from all blastocysts were extracted and chromosome aberrations were evaluated using dual-color fluorescent in situ hybridization with bovine chromosome 6- and 7-specific probes. Five embryo clone families, consisting of 112 blastocysts reconstructed from five different primary granulosa cell cultures, were examined. Overall, the mean chromosome complement within embryos was 86.9 +/- 3.7% (mean +/- SEM) diploid, 2.6 +/- 0.5% triploid, 10.0 +/- 3.1% tetraploid, and 0.5 +/- 0.2% pentaploid or greater; the vast majority (>75%) of the abnormal nuclei were tetraploid. Completely diploid and mixoploid embryos represented 22.1 +/- 4.5% and 73.7 +/- 5.5%, respectively, of all clones. Six totally polyploid blastocysts, containing or=5N chromosome complements, respectively) between two clone families were different (P <0.01), as were blastocyst yields between other clone families (P <0.01). Blastocyst yield was not correlated to % total ploidy error between clone families, but an inverse relationship (P <0.01) between blastocyst total cell number and total % chromosome abnormality was observed within embryos. Categorization of the blastocysts into three quality grades (good, medium, and poor) and comparison of the distribution of ploidies when classified into 0%, 0.1-5.0%, 5.1-10.0%, 10.1-15.0%, and 15.1-100% errors within embryos indicated that medium- and poor-grade embryos were different (P <0.05) from good-quality, in vitro-produced embryos. In a separate study, 11 different granulosa cell cultures (that did not correspond to those used for NT) were evaluated and found to possess only 0.23 +/- 0.12% ploidy errors. These results demonstrate that 1) the percentage of ploidy errors in bovine NT blastocysts is inversely related to total blastocyst cell number, 2) the mixoploid condition is representative of the majority of embryos, 3) 100% polyploid NT blastocysts can exist, and 4) the ploidy errors seem not to be derived from the donor cells.