Individual onion (Allium cepa L.) bulbs were evaluated for pungency by measurement of enzymatically produced pyruvate and by flavor perception. In four separate experiments, pyruvate values were highly and significantly correlated to mean sensory ratings. Correlation coefficients (r) were 0.92, 0.84, 0.95, and 0.79, and regression coefficients (R2) were 0.84, 0.71, 0.91, and 0.62. The high correlations indicate that pyruvate analysis can be used as a reliable selection technique for pungency in onion breeding programs.

An F2 population (Allium fistulosum x A. cepa) of 20plants, 10 BC1,[(A. fistulosum x A. cepa) x A. cepa], and 50 BC2 plants, [(A. fistulosum x A. cepa) x A. cepa] x A. cepa were studied cytogenetically and characterized for four isozyme alleles plus various morphological characteristics. All of the progenies were in A. fistulosum (the bunching onion) cytoplasm. In the F2 population we observed non-random chromosomal and allelic segregation, suppression of bulb onion allelic expression, and abnormalities in mitosis and meiosis. Most BC2 plants resembled A. cepa (the bulbing onion) morphologically, but anthers, filaments, pistils, and petals were abnormal. Only 3 plants, and these were most nearly like the F1 hybrid morphologically, produced any seeds.The data and observations support the hypothesis of nuclear-cytoplasmic incompatibility interactions between the bunching and bulb onion species.

Heritabilities of the pungency and single center traits were estimated in onion breeding populations using selection response and half-sib family analyses. Pungency was determined indirectly by measuring enzymatically produced pyruvic acid in individual bulbs. After one generation of selection, pungency was lowered by 8.1% and 8.9% in the populations 90-61-1 and 89-69-8, respectively, and realized heritabilities of 0.21 and 0.51 were estimated.Selection had no effect in lowering the pungency of population 90-62. Heritability estimates calculated through half-sib progeny analysis were 0.53, 0.48, and 0.25 for pungency in the populations 90-61-1, 90-62 and 89-69-8, respectively. The number of single centered onions was increased by 19% and 22% in populations 90-62 and 89-69-8, respectively, after one generation of selection, and the realized heritability estimates were 0.37 and 0.34, respectively.

Relatively large quantities of seed were obtained from the interspecific backcross (A. fistulosum xA. cepa) ×A. cepa allowing, for the first time, an extensive study of the heritable traits exhibited by backcross progeny. Two backcross populations, BC1034 and BC1040, distinguished by differentA. fistulosum parents, were characterized for the isozyme markersIdh-1, Adh-1, andPgi-1. Statistical methods are described to calculate cell probabilities for a mixed population of F2 and BC1 progeny, using an estimate of the fraction of F2 progeny in the population derived from the isozyme data. Cell probability distributions were calculated for a mixed population with independent pairs of loci and a mixed population with nonindependent pairs of loci. The isozyme lociIdh-1 andPgi-1 appear to be linked, with a map distance estimated at 33 centimorgans (cM) in BC1034 and 42 cM in BC1040. The probability distribution model for linked loci did not account for all of the distorted segregation ratios inIdh-1 ×Adh-1 orPgi-1 ×Adh-1. The cytological literature does not support linkage betweenIdh-1 ×Adh-1 orPgi-1 ×Adh-1. The distorted segregation ratios for these pairs of loci are likely the result of genetic incompatibilities between the two species.

No abstract