In the yeast Saccharomyces cerevisiae there are three copies of the F 1 F 0 -ATPase -subunit gene ATP1 on chromosome II (Takeda et al., 1995). However, after genome analysis using S. cerevisiae strain S288C, only one ATP1 gene sequence was observed (Feldman et al., 1994;Obermaier et al., 1995). To check whether the number of copies of ATP1 is strain-dependent or not, we carried out three different experiments: (a) long-PCR analyses of total DNAs isolated from several reference strains, carried out by preparing 29-mer oligonucleotides based on the 5 -and 3 -up-and downstream regions of the ATP1 nucleotide sequence using the data from the genome project to synthesize primers; (b) restriction analyses of chromosome II from the reference strains with SplI; and (c) long-PCR analyses of prime clones 70113 and 70804, both of which contained two ATP1 gene copies, ATP1a and ATP1b, and ATP1b and ATP1c, respectively, using 30 nucleotides just inside the 3 -end (sense) and 5 -end (antisense) of the ATP1-coding region as primers. In the case of the long-PCR experiments, the reference strains DC5, SEY2102, W303-1A, W303-1B, LL20 and DBY746, as well as strain S288C, generated a DNA fragment of approximately 32 kb, which hybridized with ATP1. During SplI digestion, a DNA fragment of more than 50 kb which hybridized with ATP1, was obtained from all reference strains. In the case of prime clone analyses using the long-PCR experiments, the distance between ATP1a and ATP1b or ATP1b and ATP1c was approximately 10 kb or 7 kb, respepectively. The S288C strain generated these two DNA fragments, as do the other strains. These results showed that all these strains contained three copies of ATP1 on chromosome II.