Premium
Androgenic Somaclonal Variation in Tobacco and Estimation of its Value as a Source of Novel Genetic Variability 1
Author(s) -
Schnell R. J.,
Wernsman E. A.
Publication year - 1986
Publication title -
crop science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.76
H-Index - 147
eISSN - 1435-0653
pISSN - 0011-183X
DOI - 10.2135/cropsci1986.0011183x002600010020x
Subject(s) - biology , population , cultivar , genetic variation , genetic variability , selection (genetic algorithm) , doubled haploidy , mating design , botany , genetics , horticulture , heterosis , ploidy , hybrid , genotype , demography , artificial intelligence , sociology , computer science , gene
In vitro anther culture of near homozygous tobacco ( Nicotiana tabacum L.) cultivars yields haploids which, when chromosome doubled, result in populations of doubled‐haploid (DH) lines exhibiting great among‐line genetic variability with agronomic characteristics inferior to those of the cultivars. Studies were conducted to characterize the nature of this “induced” genetic variability and to determine its potential value in recurrent selection programs. Sixty‐four DH lines from a single plant of cultivar ‘NC 95’ were randomly assigned to eight sets of a North Carolina Design II mating design. The DH full‐sib families and selfed progenies of the parental plant were evaluated for seven variables in two field environments. Genetic variances in the DH population were estimated, heritabilities calculated, and genetic gains from practicing half‐sib and full‐sib family selection for yield improvement in the DH population were predicted. A DH population composite, a random‐mated DH population, and populations resulting from practicing half‐sib and full‐sib family selection, in the DH population were compared to two sublines of the original parental plant. Significant additive genetic variance was detected for six of seven variables measured in the DH population. Dominance variance estimates were zero or negligible. Random mating of DH lines resulted in a recovery of 26% of the yield loss of the DH population composite compared to the cultivar. One cycle of half‐sib and full‐sib family selection resulted in populations yielding 5% and 6.0% respectively, more cured leaf than the random‐mated DH population; both selected populations were lower yielding than the sublines of the original parental plant. Two or more cycles of recurrent full‐sib selection are estimated to be required to achieve a population equal in yielding ability to the parental cultivar. This requirement reduces the appeal of androgenic somaclonal variability as a germplasm source for tobacco improvement.