The genus Zea is here divided into the Sect. LUXURIANTES Doebley & Iltis sect. n., including the perennials Z. diploperennis (2n = 20) and Z. perennis (2n = 40) and the annual Z. luxurians (2n = 20); and Sect. ZEA, including the wildZ. mays ssp.parviglumis andZ. mays ssp. mexicana (both 2n = 20), and Z. mays ssp. mays (2n = 20), the highly domesticated and tremendously variable derivate of the latter. This division is verified by a multivariate analysis of a large number of morphological characters of the male inflorescence. Cytogenetic and chemotaxonomic evidence supports the morphological conclusions. A consideration of the phylogeny of Zea within the conceptual framework offered by this new sectioning of the genus points convincingly to annual teosinte (Z. mays ssp. mexicana) as the ancestor of cultivated maize. THE GENUS Zea contains, according to its latest taxonomic treatment (Iltis & Doebley, 1980), six distinct taxa classified into four species: 1) Z. mays L., sensu lato, including Z. mays ssp. mays, the cultivated maize; Z. mays ssp. mexicana (Schrader) Iltis, and Z. mays ssp. parviglumis Iltis and Doebley, the latter two the widespread annual teosintes from Mexico and west central Guatemala; 2) Z. luxurians (Durieu and Ascherson) Bird, the annual teosinte from southeastern Guatemala and Honduras; 3) Z. perennis (Hitchc.) Reeves and Mangelsdorf, the tetraploid perennial teosinte; and, finally, 4) Z. diploperennis Iltis, Doebley and Guzmain, the recently discovered diploid perennial teosinte, the latter two both highly local species from southern Jalisco, Mexico. These taxa could be arranged subgenerically in several ways, depending on one's criteria. Thus, if one follows the only taxonomic system so far published (Post and Kuntze 1903; cf. Wilkes, 1967), the genus would be divided into Sect. ZEA, containing only the cultigenZ. mays sensu stricto, and Sect. EUCHLAENA (Schrader) Kuntze, containing all five remaining taxa, the teosintes. The reasons the genus was thus I Received for publication 19 October 1979; revision accepted 19 December 1979. Support for this work provided in part by grants (to H.H.I.) from NSF (BM S74-21861), the Research Committee of the Graduate School, the Davis Fund and the 0. N. Allen Herbarium Fund, Department of Botany, University of Wisconsin-Madison, and Pioneer Hi-Bred International, Inc. For help and support with various aspects of this research our thanks go to Donald Duvick, Robert Kowal, Walton Galinat, Frederick J. Hermann, Karen Lind, Norman Borlaug, and Linda Ainsworth of the Centro Internacional de Mejoramiento de Maiz y Trigo (CIMMYT); and especially to Maria L. Puga, Rafael Guzman M., Enrique Estrada F., and the Instituto de Botanica, Universidad de Guadalajara, Mexico. John Lonnquist grew much of our material. D. Davidse, F. G. Gould and T. Soderstrom made helpful comments on the manuscripts. divided are obvious enough. First, not only are the gigantic, polystichous, many-seeded, nondisarticulating female ears of maize totally unique in the Andropogoneae (or, for that matter, in the Gramineae), but the slender, distichous, few-seeded, disarticulating female spikes of all teosinte species are so similar to each other as to nearly defy taxonomic discrimination. Similarly with the tassel, the male inflorescence. Maize tassels are always easily recognized by their thick and highly condensed central (terminal) spike (i.e., with many spikelets per unit length), the lateral branches in contrast being always slender and uncondensed. Teosinte tassels, on the other hand, have uniformly slender and uncondensed branches, with the central (terminal) spike scarcely if at all differentiated from the lateral branches. As a matter of fact, experienced "teosinte hunters" can soon learn to distinguish teosinte and maize from a car window at quite some distance by scanning the cornfield "skyline" with binoculars and looking for the central branch of the tassel. If uncondensed and slender, it's teosinte; if stiff and thick, it's maize! Thus, simply on the basis of spikelet arrangement in both male and female inflorescences, Kuntze's taxonomic division would seem, on the face of it, to be correct and unassailable. The uniqueness of the maize ear notwithstanding, any classification of an economic genus which places the object of human desire into one subgeneric category and all its wild relatives in another is, a priori, open to suspicion. In other words, we may state as a general principle that the fundamental systematic classification of economic genera should never be based primarily on those morphological features deliberately selected for by man. To do so leads to a taxonomy where the generally largeand many-seeded cultivars are placed in