Thursday, December 2, 2010

Maize is a Machine

This is why farmers like hybrid seed. The parents on the left and right are inbred lines that have been self-pollinated for many, many years.* The two rows of much bigger plants in the middle are simply their hybrid offspring - they grow faster, produce higher yields and are tougher in the face of unfriendly environments.

Hybrid vigor (aka heterosis) is the tendency of hybrid offspring to be more vigorous than either their inbred parents or open-pollinated (OP) ancestors. The right combination of inbred lines can produce hybrid seed that produces twice the yield of naturally-crossing OP varieties. Plant breeders spend a lot of time trying to produce better and better inbred lines that have both excellent agronomic characteristics on their own and "combine" well to produce extra-vigorous hybrids. Unfortunately, about half of this extra vigor is lost in each generation that's descended from the hybrid individual. The scientific reasons for this haven't been completed settled but the economic implications are clear - the extra yield that comes from F1 hybrid seed more than covers the cost of buying new seed every year.

Northern flint, Modern, Southern dent
Prior to the 1920s, farmers grew only OP varieties. Although there are now over 100 races within the species Zea mays, Amerindians were primarily growing two when the Europeans showed up: Southern Dents and Northern Flints.** Southern dent varieties produce single stubby, white cobs on moderately tall and thick single stalks. Cobs are slow to mature and kernels are skinny, soft and develop deep dents when mature. Northern flint varieties produce multiple long, skinny, reddish cobs per stalk and plants produce multiple short, skinny stalks (tillers) per plant. Cobs mature quickly and kernels are squat and flinty when mature. European pioneers mixed up these two genotypes as they settled the Plains, creating Corn Belt Dents, and the now-familiar corn phenotype that we recognize in nearly all modern hybrid maize.

Many now-famous OP varieties were developed in the 1800s to early 1900s by American farmers: Lancaster Sure Crop, Reid Yellow Dent, Midland and Jarvis. OP variety development could be a very personal affair at this time. Reid and his dad selected their self-titled variety to have easily-plucked ears in order to spare the wrist of the aspiring artist son. Despite the care of many individual farmers, the national average yield of corn stayed very low from 1870-1920 (~ 27 bushels/acre).***

I think I've seen some variation of this graph in about half of the maize research presentations I've ever sat through. It displays the amazing increase in U.S. average corn yields since the appearance of professional plant breeders. I've been told that about 60% of this increase in yield is due to genetics (primarily the introduction of hybrid seed) and about 40% is due to agronomy (improved fertilization, tillage, herbicides and denser planting).

Modern corn is incredibly optimized to its agricultural environment. I'd be surprised if any crop has been so massively and unrecognizable modified from its wild relative into a creature so suited to our needs. It's become beneficially oblivious to the shade cast by its neighbors, holds its leaves straight at a high angle from a single stem to share light with its neighbors, and even twists while it grows to take advantage of sunlight between rows!

Despite this potential, American farmers weren't too excited about the prospect of buying new seed every year when hybrid seed was first offered in the 1920s. However, the superior performance of hybrids during the severe droughts of 1934 and 1936 led to their very rapid adoption. 1% of U.S. corn farmers planted hybrid seed in 1933, 50% in 1940 and 80% in 1944!

Since then, generations of Midwestern teenagers have earned their summer beer money de-tasseling corn. Corn is wind-pollinated, dropping pollen from its male tassels onto its (and other plants') female ear silks. If you want to produce large amounts of hybrid seed, the way to do it is plant a row of "male" plants for every couple rows of "female" plants. Female plants are simply plants that have had their tassels chopped off (hence the need for cheap teenage/immigrant labor). Seed harvested from these de-tasseled plants are only cross-pollinated, and therefore reliably hybrid.

In the 1970s, some hybrid seed producers began using de-tasseling tractor attachements to save money. An even better solution was to plant "female" lines that were genetically incapable of producing pollen, thereby completely avoiding the whole de-tasseling process. Cytoplasmic male sterility (CMS) mutations were commonly used from the 1960s through the 1980s, with infamous consequences in 1970, when a fungus discovered an unknown chink in the most popular CMS genotype's armor. Currently, the major seed companies are developing ingenious selection and sorting systems that allow plants with pollen-killing transgenes (nuclear male sterility) to produce hybrid (but non-transgenic) offspring.

Genetic male sterility is really key for hybrid corn seed producers because it saves a ton of money and oil and spares the inevitable worker injuries that occur when you work in a hot, rocky field all day, squinting into the sun while you make hundreds of knife cuts above your head. Transgenic male sterility also avoids the monoculture issue of CMS, since it works in all genetic backgrounds. Most importantly, transgenic male sterility will allow us to get the great benefits of heterosis from crops that aren't considerate enough to keep their pollen and seeds in separate flowers (rice, wheat, oilseed rape, etc.)

I began composing this post back in July as I worked my way through our tall, sticky corn fields. June bugs dripped drunkenly from emerging tassels under a darkening sky as I pawed through stiff leaves and Johnson grass. My co-workers stuck with hand shears but I quickly switched to yanking the tassels straight up out of their rosette whenever they had emerged enough to wrap my fingers around them. There was something very satisfying about hearing that wet POP! as you ripped the manhood out of yet another corn plant that stood between you and the end of the work day. Our main field would be carefully hand-crossed to make specific combinations of different genotypes, but these two side fields were just meant to produce enough hybrid seed so that the next year enough could be planted to feed to animals. 4 rows de-tasseled for ever row left intact (and unharvested). The early season weather had given us a great head start. We were pollinating well before the "go to hell date" and had high hopes for our season. Late summer smut or hail, or an early frost could still destroy it all, but so far so good.

As the welcome, cool rainshower draped over us, the irony was not lost on me of the song that had been stuck in my head all day 

"Rain makes corn..."

* the inbred lines are B73 and MO17, which are somewhat obsolete in real world ag but are established lab rats (B73 got its genome sequenced)
** Much of the information and pictures from this post were found in plant breeding lessons in the UNL's excellent Plant and Soil Sciences eLibrary  
*** Today, it's becoming common to get 300 bushels/acre with cutting edge varieties grown in the best environments. Selecting seed from the best plant in your field is actually a very inefficient way to improve your germplasm. I'll explain in a later post... (and the Corn Shows didn't help...)
 h/t CSA


  1. Great post. Just one question (prompted at least in part by Kloppenburg's First the Seed): what if some of the effort that went into producing F1 hybrids had gone into mass selection of OPs for the same sorts of traits? Might we have seen similar yield increases?

  2. That's an interesting question...

    All crops benefit from more attention from plant breeders. I guess the real question is how much of that slope is due to heterosis compared to all other genetic and agronomic improvements?

    I couldn't find a comparison of corporate inbred lines vs. corporate hybrids, but the next best comparison is probably to wheat.
    - which I expect has gotten almost as much attention from plant breeders as maize (while heterosis is not available).

    After searching around the internet for a comparison of corn and wheat historic yield, I found Steve Savage's post:

    ...which shows that wheat yield gains have been much slower, but also suggests that differences in investment and the pushing of wheat to more marginal climates may have had an influence.

    Meanwhile, soybeans (which are an industry favorite) have kept pace with corn yield gains in the past 20+ years, according to Pioneer:

    Though, they explain this by improvements in agronomy. I don't know how much investment soybeans have received in genetic improvement, though there's no heterosis and they have a infamously narrow germplasm base.

    Meanwhile, another paper concludes:

    "Finally, a strong and competitive plant breeding and seed industry, fostered by hybrid cultivars and other forms of cultivar protection, appears to be a major factor in the remarkable progress in maize yields in the United States and surely has lessons for elsewhere."

    Any crop benefits from attention from plant breeders, and many of the traits which make maize a "machine," such as changes in leaf erectness, tolerance of shade, etc. would happen with or without heterosis. However, heterosis is extremely powerful and it's hard to imagine it hasn't had a major contribution to yield. We'll see if upcoming hybrids of wheat, rice and other commodities leads to a huge increase in yield...

    *To be pedantic, "mass selection" is a technical term for walking out into the middle of your field, looking for the "best" plants and saving their seeds specifically. This is very inefficient for a lot of reasons (e.g. half the genes in that great plant's seed are those of its random neighbors and many of the "best" plants are just genetically average plants that were planted in an especially good patch of soil).

  3. Wow, Matt, that reply was almost a blog entry in itself. Thanks for the info...good stuff.

  4. haha thanks, i came across a giant paper on wheat breeding that answers a lot of these questions in more detail. so that will be on the way too...

  5. I'm enjoying your posts, as usual! I would love to hear your thoughts on this, if you are looking for a topic: Thanks, and Merry Christmas. :)

  6. This comment has been removed by the author.

  7. I stumbled on this blog and did not read the whole thing but you are incorrect regarding Northern Flints and Southern Dents. The long slender ear is from a Northern Flint and the ear with many kernel rows is a Southern Dent.

  8. Thanks for the correction, William.

    I rechecked my source and did mix it up



Related Posts with Thumbnails