V. corymbosum L. selections and hybrids; 4) southern highbush blueberries, which are basically the same as the more common highbush types, with the addition of genes from the low-chilling species V. darrowii Camp, as well as from V. angustifolium Ait., V. ashei Reade, and V. tenellum Ait. (Ballington et al., 1991a; Lyrene, 1990a); and 5) rabbiteye blueberries, which are all wild selections or hybrids derived from V. ashei. In general, current breeding objectives for the numerous blueberry breeding programs include: 1) expansion of the range of production through broader soil adaptation (upland mineral soils); 2) adaptation to lower chill regions or colder production regions through increased bud tolerance or delayed flowering combined with earlier fruit maturation; and 3) increased disease resistance. Genetic fingerprinting, gene tagging, and mapping Rabbiteye blueberry cultivars constitute a very narrow germplasm base, which has been recognized as a problem (Lyrene, 1981b). To determine the magnitude of the problem and devise a method to widen the germplasm base, Aruna et al. (1993) utilized 20 randomly amplified, polymorphic DNA (RAPD) primers to determine genetic relatedness among four native clones and 15 improved cultivars of rabbiteye blueberry. In an extension of the previous study, a cultivar key for economically important rabbiteye blueberry was created using 11 DNA fragments amplified by four RAPD primers (Aruna et al., 1995). Clustering of genotypes generally made sense in light of known pedigree information. Overall, there was a trend toward lower genetic distances within advanced generations from the wild selections. Levi and Rowland (1997) utilized 15 RAPD markers and three simple-sequence repeat (SSR)-anchored primers to differentiate 15 highbush blueberry and two rabbiteye cultivars and one southern lowbush selection from the wild. However, the molecular data showed clustering within V. corymbosum that did not agree with pedigree information. In a similar study, 26 wild lowbush blueberry clones (V. angustifolium), including six named cultivars and 12 selections, were screened with a total of 30 RAPD primers (Burgher et al., 1998). All 26 clones could be unambiguously identified with 11 of the primers, and clusters matched known relationships among the clones fairly well, showing trends relating to geographic origin. Several efforts are under way to create genetic linkage maps for cultivated blueberry in order to tag quantitative trait loci (QTLs) related to chilling requirement, cold hardiness, and soil adaptation. Rowland and Levi (1994) constructed a RAPD-based genetic linkage map of diploid blueberry based on a cross between an F 1 hybrid created between V. darrowi x V. elliottii Chapm. and another V. darrowi genotype. The map was constructed from over 70 markers and consisted of 12 linkage groups, which represented the basic chromosome number for blueberry. More recently, Rowland et al. (1999) developed RAPD-based genetic linkage maps based on crosses between F 1 hybrids of V. darrowi and V. c...