Maize Floral Development: New Genes and Old Mutants.

Abstract

Although transgenic approaches toward plant improvement have begun to yield some modest results, one of the more spectacular achievements in the history of agriculture was attained more simply. According to the most widely held view, maize, with its kernel-laden ears, was derived in the last 10,000 years from teosinte, a grass having only marginal agronomic value (Beadle, 1980; Galinat, 1988). Were the maize ear a more recent invention, its elegant structure would be the subject of several patents. Its multiranked array of kernels presents a great improvement in packing efficiency over the distichous arrangement of teosinte. These tightly packed kernels, instead of being encased individually, as in teosinte, are exposed and free of chaffy bracts. Finally, this dense array of nutritious endosperm is conveniently packaged in a set of protective, but easily removed, husk leaves. How the maize ear might have been engineered by new world farmers who possessed only a rudimentary knowledge of hereditary principles has been a source of great debate (Galinat, 1983; Iltis, 1983), much of it framed in various interpretations of fossil (Mangelsdorf et al., 1967) and anatomical (Doebley and Iltis, 1980; lltis and Doebley, 1980) data. Lying at the root of this controversy is our limited knowledge of how morphogenesis is controlled in plants. Although no definitive answer as to the origin of the maize ear can be offered here, we can try to outline the logic that guides its development. We begin by describing the sequence of events by which the ear normally develops and how this basic pattern is modified to produce the staminate tassel. We then discuss the results of genetic analyses, which suggest that inflorescence development can be modeled as a process involving several semiautonomous programs. Finally, we describe how these formal models of gene interaction can be tested and refined through a variety of molecular approaches.