ABSTRACT In the paper by Jules O'Rear and Jasper Rine (Genetics 113: 517-529; July, 1986) entitled "Precocious meiotic centromere separation of a novel yeast chromosome," the authors described a gene conversion event between a linear yeast plasmid carrying a LYS2 gene and a mutant lys2 gene at the wild-type locus on chromosome II. When these yeasts were mated to wild-type yeast and the resulting diploids sporulated, linked markers on the linear plasmid showed unusual segregation and poor spore viability was observed. On the basis of these observations, we proposed that the recombination event between the linear plasmid and chromosome II had split chromosome II into two fragments, one of which carried the normal centromere of chromosome II (fragment IIa) and the other, a telocentric fragment (fragment IIb), carried the centromere present on the linear plasmid. Separation of the chromosomes from these cells on OFAGE gels verified that chromosome II had been split into two fragments. Furthermore, we proposed that the sister chromatids of the telocentric fragment (fragment IIb) separated precociously in meiosis I when complete chromosome II and fragment IIa were present. In discussions with colleagues, an alternative explanation arose in which a recombination event between a sister chromatid of fragment IIa and a sister chromatid of chromosome II would result in each chromosome II chromatid being joined to a fragment IIa chromatid at CEN2. The two daughter cells of meiosis I would therefore each receive one chromatid of fragment IIa and one chromatid of chromosome II. Segregation of the two sister chromatids of fragment IIb to one pole in meiosis I without precocious centromere separation would result in the observed tetrad classes. To distinguish between these two mechanisms, a centromere-linked marker was introduced into the cross between the strain containing the two fragments of chromosome II and a wild-type strain. Tetrad analysis of the resulting diploid is consistent with the recombination model for the poor spore viability and inconsistent with precocious centromere separation. We thank Drs. Eric Lambie, Michael Lichten and Tom Petes for helpful discussions.