Specialized traits of the oldest known maize cobs are absent in the teosinte tassel (contrary to basis of CSTT)

According to the CSTT of Iltis (Science 222:886-894, 1983), the maize ear originated by a transmutation in teosinte of the target area for expression of the secondary male traits to also include the female spike. This transmutation supposedly accounts for the archaeological record in which there seems to be a sudden despecialization of the teosinte female spike with a concomitant loss of induration, a reduction of cupules and a reactivation of the second member of paired spikelets. But the fact is that the oldest maize cobs are still more similar to the teosinte ear than to its tassel in regard to both cupule development and glume shape.

The phenotype of the tassel seed mutants of maize should be an example of what happened according to the CSTT. But the morphology of neither tassel seed maize nor of tassel seed teosinte fits the Iltis theory. In both cases female development in the tassel is associated with the fruit case derivatives for cupules, induration and glume shape typical of their normal target areas. Even with normal teosinte and maize, the female areas within the mixed (bisexual) inflorescences that usually terminate tillers are also associated with these fruit case derivatives as expressed in their ear type, contrary to the expectations of CSTT.

It is also significant that the long rachilla, that is characteristic of the oldest cobs is absent in both the ear and tassel of teosinte and so it could not be derived from transmutation from the tassel. Rather the long rachilla is one of several pleiotropic effects that are controlled primarily by a series of multiple alleles at the tunicate locus. These other effects include softer, longer female glumes and some cupule reduction, which are traits that Iltis would attribute to transmutation, but more probably came from domestic selection for a tunicate allele.

The effects of a weak tunicate allele still occur in Chapalote, an ancient indigenous race of maize in Mexico. Its phenotype in Chapalote is considered by Mangelsdorf to be similar to that of its prehistoric precursor in the oldest cobs from Bat Cave, New Mexico and probably to that of certain of the oldest known cobs from caves near Tehuacan, Mexico. Since Iltis ignores any role for the tunicate locus during the origin of maize, the domestic advantages of an elongate rachilla, in the ear need to be mentioned here. The long rachilla not only elevates the kernel to a level near the apex of the glumes and, thereby, exposes the kernel except for a little membranaceous tissue from the lemmas and paleas, but it also reflexes the spikelet away from the cupule and thereby makes it threshable. Protection from birds at this stage becomes dependent upon an enclosure of the entire ear by husk leaves borne on the shank.

Other serious errors of fact in the Iltis article include the statement (p. 888) "no key genes differentiating maize from teosinte have ever been found." He goes on to explain, "This is because, in fact, they do not as such exist." The fact is that genes controlling the key traits are known: (Pd pd) - paired vs. single female spikelets and (Tr tr) - many vs. two-ranked spike. All combinations of these genes occur in the F2; although the phenotypes of the teosinte alleles tend to be unstable in a highly evolved maize background, their expression can be stabilized through selection. As a single event in the origin of maize, the transmutation speculated by Iltis should segregate as a simple Mendelian factor in the F2 of hybrids and, thereby, yield only maize and teosinte. This is obviously not the case.

The morphology of the oldest maize cobs and the genetics of maize-teosinte hybrids make the contentions of Iltis untenable.

Walton C. Galinat

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

Return to the MNL 58 On-Line Index
Return to the Maize Newsletter Index
Return to the Maize Genome Database Page