Microsatellites, tandem arrays of short (2-5 bp) nucleotide motifs, are present in high numbers in most eukaryotic genomes. We have characterized the physical distribution of microsatellites on chromosomes of sugar beet (Beta vulgaris L.). Each microsatellite sequence shows a characteristic genomic distribution and motif-dependent dispersion, with site-specific amplification on one to seven pairs of centromeres or intercalary chromosomal regions and weaker, dispersed hybridization along chromosomes. Exclusion of some microsatellites from 18S-5.8S-25S rRNA gene sites, centromeres, and intercalary sites was observed. In-gel and in situ hybridization patterns are correlated, with highly repeated restriction fragments indicating major centromeric sites ofmicrosatellite arrays. The results have implications for genome evolution and the suitability of particular microsatellite markers for genetic mapping and genome analysis.Runs of repetitions of short sequence motifs 2-5 bp long, described as microsatellites or simple sequence repeats, are probably ubiquitous elements of eukaryotic genomes (1-3). They provide highly informative and polymorphic markers for genetics (4-6) or plant, fungal, and animal fingerprinting (7,8). Genetic mapping using microsatellites as markers involves amplification of repeat arrays using PCR with primers flanking the arrays. Primers are often chosen based on database searches or sequence data from cloned arrays (9-12), whichgive selective data about genomic distribution. However, little is known about the real chromosomal organization and physical localization of microsatellite motifs within plant genomes. The only microsatellite repeat mapped physically in plants, the polypurine motif (GAA)7, has been correlated with the positions of C-bands in barley (13). Some physical mapping in fish and primates (8, 14) using in situ hybridization shows clustering of microsatellites on some chromosomes. Recent data in mouse show that mapped microsatellites, mostly (CA)n repeats mapped by PCR polymorphisms, are distributed among autosomes in proportion to chromosome length, while the X chromosome shows a clear deficit of the microsatellites (15). Within chromosomes, the frequency of both large and small clusters with respect to meiotic crossovers slightly exceeded expectation.Sugar beet (Beta vulgaris L.; 2n = 2x = 18) is a valuable model species for investigating the large scale organization of the nuclear genome because (i) the genome is relatively small with 758 Mbp (16), (ii) fluorescent in situ hybridization can accurately locate sequences along the metaphase chromosomes and within interphase nuclei (17), (iii) major classes of the repetitive DNA have been characterized including both satellite and retrotransposon sequences (18)(19)(20)(21)(22), and (iv) microsatellites are known to be highly abundant (23).Here, we aimed to characterize the genomic distribution of microsatellite sequences in sugar beet. We used seven simple sequences representing a range of nucleotide motifs, many used for g...