Polymitotic, canalization and probability

The maize mutant polymitotic (G.W. Beadle, Science 70:406, 1929, Cornell Ag. Exp. Stat. Mem. #135, 1932; West, M.S. thesis, 1984, Ph.D. thesis, Univ. of Minn., 1985) reveals a locus involved in the orchestration of the developmental steps of meiosis. Specifically, this locus is critical to the cessation of this developmental subroutine. Mutation at this locus results in the failure of meiosis II to conclude. Male sterility is complete, but female sterility only partial.

Histological analysis of polymitotic megasporogenous cells reveals telophase II chalazal cells either degenerating from polymitoses or proceeding to a normal conclusion of meiosis. These two options are reflected in the interspersion of ovules containing megacytes and ovules containing normal embryo sacs on ears of polymitotic homozygotes, the former causing female sterility and the latter giving rise to randomly scattered, normal, viable seed (West, 1985).

Mutation at polymitotic reveals a genetic switch operative in the control or canalization (C.H. Waddington, The Strategy of the Genes, 1957; J.M. Rendel, Canalization and Gene Control, 1967) of development. This is an example of a class of mutants that decanalize development and in so doing reveal an "underlying genotype" (A.S. Fraser, Genetics 57:919, 1967; see also B.T.O. Lee and P.A. Parsons, Biol. Rev. 43:139, 1968, and J.N. Thompson, Stadler Symp. 9:63, 1977). The canalized mode of expression is characterized by dominance (with some exceptions), regulation and very low rate of failure in the orchestration of developmental events. Canalized development gives rise to traits that distinguish species. The decanalized mode is probabilistic, exhibits genetic variance and selection response.

In the case of polymitotic, at the end of meiosis II, each ovule on any given ear faces the same probability of polymitoses. Those not so afflicted are lucky, not fitter. The probability of normal embryo sac formation on mutant plants is a heritable trait of the ear-bearing sporophyte. The number of lucky embryo sacs (and eventual seed) borne on an ear of a polymitotic homozygote is a direct reflection of the genotypic probability value (or potential) of the individual sporophyte. These values vary widely among different genotypes and can be analyzed in the standard quantitative genetic manner, using

seed-set as the metric. Seed-set levels are easily modified by selection (West, 1985). The genotypic value altered by selection is a probability. This suggests that the non-mutant allele canalizes development by raising the probability at this genetical switch to 1 (or nearly 1) for the proper conclusion of meiosis.

For polymitotic, canalization, dominance and regulation appear as different aspects of a single mode of genetic function. The canalized mode contrasts an alternative mode which is probabilistic and tractable to selection. This view is consistent with observations of several mutants in Drosophila that affect the canalized development of body hairs and wing veins (cf. Waddington, Lee and Parsons, Rendel, Thompson, op.cit.). However, the probabilistic nature of the decanalized mode would not be recognized with mutations which decanalize traits manifest in unreplicated organs or tissues, except as variance in penetrance or expressivity of the decanalizing mutation, or non-correspondence between genotypic and phenotypic segregations (opaque-7? cf. K.S. McWhirter and R.A. Brink, in Walden, ed., Maize Breeding and Genetics, p. 373, 1978). The high level of replication of a single structure and event (the ovule and megasporogenesis) on the maize inflorescence provides a unique opportunity for observing this phenomenon in polymitotic homozygotes.

A mechanism by which natural and artificial selection modify the level, extent or timing of expression of a trait is suggested by the alterability by selection of the probability of alternative states of a genetic switch. Whether such changes in probabilities are a consequence of alleles at multiple loci (D.S. Falconer, Intro. to Quant. Genetics, 1981), or to potentiation of a single functional unit at different levels (W.E. Castle and J.C. Phillips, Carnegie Inst. Wash. Pub. #114, 1914) has not been determined.

David P. West

Please Note: Notes submitted to the Maize Genetics Cooperation Newsletter may be cited only with consent of the authors.

Return to the MNL 59 On-Line Index
Return to the Maize Newsletter Index
Return to the Maize Genome Database Page