2. Experiments about the Origin of Corn

A - Native Indian Corn: We were able to obtain through the help of Brazilian colleagues, of Dr. Cardenas of Cochabamba and of Dr. Cutler, authentic "wild Indian" corn. The Bolivia corn from Cochabamba grew very well at the low altitude of Piracicaba, flowered generally well, but produced many poor ears. Material from the lowlands of Mato Grosso (Brazil), from Paraguay and the Bolivian Chaco is much more satisfactory. But in nearly all cases it was rather difficult to maintain the strains, since they degenerate very rapidly with more or less close inbreeding. The following material has been studied genetically.

"Acre" from the territory of Acre (Brazil). The plants are very tall without tillers, ears long and slender with 8 rows, grains large, round and soft, exhibiting the following colors: dominant purple (ACR Pr), red (pr pr) or recessive colorless (probably rr), brown aleurone (lost), yellow or white endosperm.

"Chavantes" (from the State of Mato Grosso, Brazil). Very tall plants, segregating semi-dwarf, ears big and heavy, 12 or more rows, grains large, soft, white or sometimes tinged, purple (Pr), red (pr pr) or light pink (pericarp?). The constitution of these grains is probably AA CiCi RR as shown by the following test cross with C sh: (F₂):

The dominant inhibitor Cⁱ is not completely dominant and varying percentages of the kernels with the constitution Cⁱ-Sh- are not white, but very pale purple and red. It seems as a whole that the Indians selected modifiers which reduce all possible color in the kernels as much as possible.

White endosperm is only incompletely recessive to yellow and there is present some kind of pericarp color which however becomes clearly visible only after outcrossing.

"Diamantino" (from Mato Grosso, Brazil). We received three lots of seeds. In all of them the ears originally were heavy and many rowed. The color of grains varied.

Diamantino I, had deep red pericarp (P) segregating normally after crossing.

Diamantino II had dirty brownish-orange kernels, due to orange, white or colorless pericarp on yellow-orange endosperm and sometimes yellow-brown aleurone. The segregation for pericarp color was interesting in so far as its existence could be verified only in some years, and in one year only classification between orange and colorless pericarp was very easy. In this year orange pericarp was in some instances so intense as to give a bright red color.

Diamantino III contained colored and colorless aleurone over orange endosperm, sometimes covered by orange pericarp (white cob). Absence of aleurone color may be due either to a dominant or recessive inhibitor. The former is certainly an allele to the C factor as shown by the linkage test with CC sh sh. But there are a large number of modifiers acting and disturbing the ratios. The ears collected after selfing fell into two groups. In the first there was an excess of colorless-shrunken grains combined with a deficiency of the colored-shrunken grains. In the other group of ears, besides this deviation, there appeared a deficiency in the number of the normal grains and a corresponding excess in the colored-shrunken grains.

Plant color in most strains of all three forms of native corn, Acre, Chavantes and Diamantino, is either dilute purple or dilute sun red. But the culm is very frequently heavily colored, and this color seems, at least partially, independent from A-B-Pl mechanisms.

In the shucks various colors were observed which may be either "sun red", deep purple, dilute purple, red and reddish-brown.

Finally, the glumes and the whole base of the grains may be deep or light purple or red, independent from cob color. Apparently somehow this color depends upon the same factors as the color of the shucks.

So far the existence of these different colors in vegetative organs has been registered; but it has not yet been possible, owing to lack of time, to start on a detailed genetic analysis.

If we take all characters into consideration, it seems that the indigenous strains from Mato Grosso together with the material collected by Cutler in Paraguay and the Bolivian lowlands form a natural group. Similar traits may be found also in local forms, cultivated in São Paulo. In all of them there appears, with more or less frequency, all or some of the following characters.

Slender and long ears with flexible rachis. Grains half covered by their glumes. Kernels more or less round or pointed, containing soft starch. Anthocyanin generally absent in the aleurone owing to the presence of inhibitors at the C-locus. On the other side there is a tendency for the appearance of brownish-orange colors, in the aleurone, endosperm, and pericarp.

Three characters seem to me especially important: the brownish-orange color of the kernels which may be considered as an approximation to a natural "wild" color, the slender and flexible rachis and the development of large glumes which may be taken as a change in the direction of pod corn. Their widespread occurrence can hardly be considered as a coincidence, in view of the old hypothesis, recently taken up again by Mangelsdorf and Reeves, that pod corn is the most primitive of all the different types of maize and that the lowlands on both sides of the Rio Paraguay, i.e., the triangle formed by lower Bolivia, western Mato Grosso and Paraguay, may be the geographic centre of the origin of maize. On the contrary, I think our observations, very briefly reported above, support strongly this hypothesis.

F. G. Brieger