How Did a Wild Mexican Grass With a Few Hard-Cased Kernels Become Corn?

Corn’s Wild Parent Is a Branching Grass Called Teosinte

Corn’s wild ancestor is a grass called teosinte (say tay-oh-SIN-tee). The type most closely tied to corn is Zea mays subspecies parviglumis, which still grows in the seasonally dry tropical forest of the central Balsas River region in southwestern Mexico [2].

Compared with corn, teosinte looks like a different kind of plant. It is highly branched, while corn has one main stem with short side branches that each end in a single ear [1]. Its ears are tiny and carry few kernels, in two rows, and every kernel sits in its own hard case [1] [3]. Corn kernels are “naked,” with no such case, and a corn cob carries 8 to 22 rows [1].

So the question is not only where corn began. It is also how a plant could change this much, and how we know.

Genetic Tests Point to One Start in Southern Mexico About 9,000 Years Ago

One way to find where a crop began is to compare its DNA with wild relatives. A team led by Yoshihiro Matsuoka and John Doebley tested 264 plants, each at 99 spots in the DNA, covering maize and teosinte from across the Americas. They concluded that all maize came from a single domestication in southern Mexico about 9,000 years ago. Their estimate was 9,188 years before the present, with a very wide margin of error (roughly 5,700 to 13,100 years), so “about 9,000” is a rough date, not a birthday [4]. The wild plants closest to maize came from the central Balsas River area [4].

Digging gives a second, independent clue. At Xihuatoxtla Shelter in the Balsas valley, researchers looked at stone tools, including grinding stones, from layers dated to roughly 8,990 to 8,610 years ago, and at older layers below. All the tools they tested from those levels carried maize starch grains, and tiny glassy plant bits called phytoliths showed up in the sediment as well [2]. That tells us people were using maize that long ago. It does not tell us what the ears looked like, because no cobs were reported from this evidence.

The people who did this left no names. What we can say is that human choices, repeated over many generations, are what changed the plant [9].

A Gene Called tga1 Helps Open the Stone Case

Genetics can tell us what changed, not only when. In 1993 a team led by Jane Dorweiler reported that a spot in the DNA called teosinte glume architecture 1 (tga1 for short) largely controls the difference between teosinte’s hard-cased kernels and maize’s exposed ones [5].

A 2005 study then narrowed it down. It traced the difference to a tiny stretch of DNA, only about 1,000 letters long, where maize and teosinte differ in just seven places. One of those seven changes alters the protein the gene makes [6]. So a very small change in one gene can go a long way toward turning a stone-cased kernel into a bare one. That does not mean one gene made corn. The rest of the plant changed too.

A Gene Called tb1 Helps Turn a Bushy Plant Into One Main Stem

The other big change is the shape of the plant. Biologists call a plant's habit of growing one main stem and holding back its side branches apical dominance. In 1997 Doebley and colleagues found the gene teosinte branched1 (tb1) and showed that it helps hold back side branches. The maize version of the gene is switched on at about twice the level of the teosinte version [7].

This is a useful detail. The difference did not come from a brand-new gene. It came from the same gene being used more strongly, which suggests that changes in how much a gene is used can matter as much as changes in the gene itself [7].

Ancient Cobs Show Corn Was Still Small and Unfinished Thousands of Years Later

Genes tell us what changed. Real cobs show how far along the plant was. Two finds, both from Mexico, are worth knowing.

In Guilá Naquitz Cave in Oaxaca, researchers dated cobs directly and found they are about 6,250 calendar years old, the oldest cobs known from the Americas when the study came out in 2001. They were small and primitive-looking, one only about 2.5 centimeters (1 inch) long, and botanists who examined them thought they were either hybrids or primitive maize with strong teosinte influence [8]. Even so, they were already domesticated, so people had been shaping the plant for some time [8]. The authors were careful to say the cave cannot show where the farming practices began [8].

A second cob, from the Tehuacán Valley, is about 5,310 years old, roughly halfway between the start of domestication and today. In 2016 a team sequenced its DNA. It was less than two centimeters (under an inch) long, with just eight rows of kernels, yet its genes were on the whole closer to modern maize than to teosinte [9]. Many key genes had already been changed by human selection, including the loss of the hard seed casing and changes in flowering time. Others had not: the sugar content of the kernels, and a gene linked to how kernels scatter from the plant [9].

These two cobs are different individual plants from different places, so it would be wrong to read a smooth “getting bigger” line from them. What they do show is that the change was slow and uneven: some traits arrived early, and others took thousands of years.

The Story Did Not Finish in One Place

The genetic tests above say corn began once. A 2018 study of maize genomes, including ancient ones, suggests the story then spread. It found that partly domesticated maize left Mexico and was in South America by about 6,500 years ago, and that the southwestern Amazon may have been a second place where the crop kept being improved [10] [3]. In this picture maize was not a finished product that traveled. It kept changing under people’s care as it went.

These two ideas do not fight each other. One is about where the first domestication happened. The other is about where it was completed.

Try It at Home: Count the Rows on a Cob

If you can get an ear of corn, you can compare it with the ancient evidence. This is your own experiment, not something from the sources.

1. Take a cooked or raw ear of corn. Count the rows of kernels running along its length. Is the number always even?
2. Compare. The Tehuacán cob had 8 rows. Teosinte has 2. Corn cobs are described as having 8 to 22 [1].
3. If you have a few ears, compare them. Do they all have the same number of rows?

If you want to read more, the 2001 Guilá Naquitz paper and the 2018 South America study are good places to start, and the maize pages at Earth@Home have photographs of teosinte next to maize.

Sources

Update history