Previous studies of peach germplasm using pedigree information and isozyme polymorphism data have shown limited diversity in the U.S. gene pool. To further investigate the genetic diversity among peach cultivars grown in different regions of the United States, 94 RAPD markers were used to estimate the genetic distances among 136 cultivars. Of the 12 clusters formed in a dendrogram, the 90 U.S. cultivars and breeding lines and most of those from Europe and Latin America grouped to only three clusters, while the 23 peach entries from India, Pakistan, Russia, Okinawa, and China, as well as the almond cultivar used as an outgroup, were distributed among the other nine clusters. Therefore, the genetic diversity within temperate U.S. peach germplasm is quite limited, and to expand the variability, additional germplasm should be obtained, especially from Asia. Comparison of genetic similarity based on inbreeding coefficients with similarity coefficients based on the RAPD data produced a correlation of 0.395, which is comparable to values in similar investigations in other crops. Thus, similar conclusions can be drawn from these two sources of information. RAPD data are useful particularly when pedigree information is incomplete, there has been substantial selection within breeding populations, and a high proportion of alleles are identical in state but not by descent.
The absence of seed lectin in common bean (Phaseolus vulgaris L.) was shown to be inherited as a single recessive gene and allelic to genes conditioning 6 different lectin types. In inbred backcross lines, the allele from ‘Sanilac’ (a navy class bean) for the presence of lectin was semidominant to the lectinless allele from ‘U.I. 1140’ (a Great Northern class bean) for quantity of lectin. Backcross lines with lectin (L/L) and without lectin (1/1) were developed using 2 lectinless donor parents (‘U.I. 1140’ and ‘U.I. 111', a Pinto class bean) and ‘Sanilac’ as the recurrent parent. Backcross lines and parents were grown in the field and analyzed for days to flower, seed yield, seed weight, percentage protein, and quantities of lectin and phaseolin. There were significant differences between lectin genotypes (L/L vs. 1/1) for all traits except yield, seed weight, and nonphaseolin nonlectin protein. Backcross lines without lectin had substantially higher levels of phaseolin and slightly more total protein than lines with lectin. The data suggested that phaseolin over-compensated for the absence of lectin in 1/1 backcross lines. The implications of these findings toward the nutritional improvement of bean protein are discussed.