We crossed Zea diploperennis with a primitive popcorn race, Palomero Toluqueno, in a small isolated garden in Chapel Hill in the early summer of 1979. The teosinte proved not to carry the Ga cross-sterility factor and no other barriers to crossing were apparent. We obtained abundant crossed seed.
In order to obtain F1 and F2 populations as rapidly as possible, we arranged for two winter plantings of the F1 plants to be made in Homestead, Florida, and two additional ones in Argentina, in case the Florida plantings should be lost by an early winter freeze.
All four of the F1 plantings succeeded and from two of them, one in Florida and one in Argentina, seed was harvested for F2 plantings. The results of the Argentine plantings are reported in earlier pages of this Newsletter (see item from Buenos Aires). This report is concerned only with plantings made in the United States.
F2 populations in Florida: We grew two F2 populations in Homestead, Florida, one in the experimental plots of Pioneer Hi-Bred; the other in the fields of Agricultural Alumni Seed Improvement. For the special attention that they gave our plantings, we are indebted to Dean Wrucke of Pioneer Hi-Bred and Karalee Holden of Ag. Alumni.
The principal purpose of the Florida plantings was to get some idea of the variation that we might expect in F2 populations and to this extent, they were successful.
It had been suggested to us by several skeptics that the inheritance of growth habit would prove to be so complex that, combined with the segregation for the botanical characteristics distinguishing corn and teosinte, we might expect to find annual teosintes, if they occurred at all, in such a low frequency that populations of many thousands of plants would be required to discover them.
The opposite proved to be true. In both of the Florida plantings, comprising only several hundred plants, a number of annual-teosinte phenotypes occurred. We photographed these as a matter of record. The Florida plantings were the first to show us that the inheritance of annual vs. perennial growth habit is not at all complex.
F2 populations in Texas: The Texas planting was made at a field station of Texas A and M University located in the Brazos River Bottom about 15 miles west of the city of Bryan. The main purpose of this planting was to grow an F2 population large enough to not only obtain numerous annual-teosinte phenotypes, but also to compare these with plantings of the presently-known races of annual teosinte and also to obtain photographs of such comparisons. Another purpose was to study the inheritance of the contrasting characteristics involved in the cross and to detect genetic linkages, if present.
An initial planting of about 4000 plants made in April was a total failure because of unprecedented hot, dry weather. Although irrigation was available and freely used, the F2 plants did not thrive, apparently suffering from the excessive heat. The first planting was useful only in showing that F2 populations derived from high-altitude parents will not ordinarily flourish under the hot, dry conditions that prevailed in the summer of 1980.
The second planting made in early August fared little better. It not only suffered from the continuing heat and drought but was also subject to the attacks of a disease, not ordinarily occurring on corn, as well as insect damage. Despite the application of pesticides, many plants were lost.
Among the plants that survived and finally flowered in late October, there were numerous annual-type plants with teosinte-type pistillate inflorescences. Thus, this planting agrees with F2 populations grown in Florida and Argentina in showing that the recombinant annual-teosinte phenotypes are not rare and that the number of gene pairs involved in the inheritance of the characteristics under study cannot be great.
A backcross population: Somewhat more successful was a small backcross
population of the diploid perennial parent by the F1. This comprised 147
plants. We classified these into five categories with the following results:
| Category | No. of Plants |
| Perennial | 2 |
| Weakly Perennial | 58 |
| Intermediate | 66 |
| Resembling F1 | 16 |
| Duplicating F1 | 5 |
The first and last data are interesting in showing that the spectrum of variation in this population ranges between plants almost identical to the perennial parent and plants resembling the F1 hybrids. This has to be theoretically expected. No annual- or perennial-corn phenotypes were to be expected in this backcross population and none occurred.
If we count the plants classified as "intermediate" as annuals, as they probably are, since they lack rhizomes, the ratio of perennial to annual phenotypes is 60:87. This is not a highly significant deviation from a 1:1 ratio theoretically expected if one major gene pair is involved. The data at least fit a 1:1 ratio somewhat better than they fit the 3:1 ratio that would theoretically be expected if two gene pairs were involved in the inheritance of annual vs. perennial growth habit. Another possibility that should perhaps be considered is that the cross may involve not one, but two genes for annual growth habit carried by the same chromosome.
This backcross population is most interesting because it may duplicate the manner in which annual-teosinte types arose in Mexico centuries ago as the progeny of natural crossing of Zea diploperennis and a cultivated corn in an early stage of domestication. The F1 hybrids that occurred in colonies of perennial teosinte would have pollinated the perennial teosinte and in subsequent generations, the colony would have contained both perennial and annual teosintes but no perennial corns.
All of the results of the Texas plantings are consistent with the Wilkes hypothesis on the origin of annual teosintes.
Paul C. Mangelsdorf, Lewis M. Roberts and John S. Rogers
Return to the MNL 55 On-Line Index
Return to the Maize Newsletter Index
Return to the MaizeGDB Homepage