The mechanism of the mixed inheritance of chloroplast genes in Pelargonium. Evidence from gene frequency distributions among the progeny of crosses

• Published in: Theoretical and applied genetics 1981, Vol.60 (1), p.43-53
• Main Authors: Tinley-Basset, R.A.E. (University Coll. of Swansea (UK). Dept. of Genetics), Birky, C.W
• Format: Article
• Language: English
• Subjects: Pelargonium
• ISSN: 0040-5752; 1432-2242
• DOI: 10.1007/BF00275177

The distributions are given of gene frequencies among embryos after G X W and W X G plastid crosses within and between eight Pelargonium cultivars and some of their inbred or hybrid derivatives.Two distinct segregation patterns are recognized. Homozygous type I female parents (Pr1Pr1) have a high frequency of progeny with only maternal alleles, are intermediate for biparental and low for paternal offspring. Heterozygous type II female plants (Pr1Pr2) have an equally high frequency of maternal and paternal offspring and a generally low biparental frequency. These correspond to L-shaped and U-shaped gene frequency distributions respectively in which the only modes are at 0 per cent (maternal embryos) and 100 per cent (paternal embryos), with no mode corresponding to the population mean and no sign of a Gaussian distribution.The extremely variable plastid gene frequencies are strongly influenced by the maternal nuclear genotype and by the plastid genotype in which the wild-type allele is always more successful than the mutant in strict comparisons.The relative frequencies of maternal and paternal zygotes, and the mean gene frequency among all the zygotes in a cross, are explicable in terms of the input frequencies of genes from the two parents, their degree of mixing, and by some form of selective replication of plastids. This selection is controlled by nuclear and plastid genotypes which may act in the same direction, to increase the frequency of either the maternal or the paternal alleles, or in opposition. But selection alone is inadequate to explain the shapes of the gene frequency distributions. Instead, a model is proposed in which the segregation or replication of plastids appears to have a strong random element, which results in random drift of gene frequencies within a heteroplasmic zygote or embryo.