The Secret Larder of Ancient America

There’s a fascinating paper in Nature Plants by Mueller et al. Growing the lost crops of eastern North America’s original agricultural system. When you think of Native American agriculture you probably think of maize, beans and pumpkin. What Mueller et al. look at is a much earlier agricultural system. The plants grown in eastern North America were things like sumpweed, maygrass, goosefoot, erect knotweed and little barley.

If you’re not familiar with these crops, then it’s understandable. By the time the Europeans arrived this had been replaced with the better-known crops. This is what we see in the historical, ethnographic and archaeological record of the time. To find this early Native American agriculture, you have to go much earlier where the evidence is purely archaeological and botanical.

The earliest records that the native Americans were growing something else comes from structured deposits of seeds. Seeds fall into soil, but they don’t fall en masse into pits. Neither do they accidentally fall into bags and baskets, so we know they were deliberately collected. If you’re sceptical, you could ask how do we know they were eaten? There’s been someone sifting through human palaeofaeces, and among everything else, they found the remains of seeds.

But maybe they were just spectacularly good at gathering, how do we know it was agriculture? For that, there’s the botanical evidence. You can measure differences in the morphology of the seeds. They use Chenopodium berlandieri, goosefoot, as an example. The domesticated form of C. berlandieri has a much thinner seed coat, than its wild relative and experiments with modern domesticated goosefoot might explain why.

Mueller et al. refer back to Chenopodium berlandieri ssp. nutalliae, domesticated in Mexico, to show the importance of tending the crop. They find that this form of C. berlandieri will not germinate until it’s been in cold moist soil for a few weeks. Then it will germinate easily. Because I can be a bit slow of thinking, I thought that ease of germination would make it much stronger as a weed. Seed that germinates well surely has an advantage. It turns out the opposite is true.

It’s not just a matter of germinating, but also when the seed germinates. The plant comes to seed in autumn. If the seeds scatter and plant then, they will get cold moist soil and germinate. However, they’ll be germinating into a winter, when they will be too cold and die. These seeds need to be stored in dry conditions away from the soil, so they don’t germinate, and are ready for early Spring when the conditions are right and they’re growing into Summer.

They also discuss why these plants were domesticated. Over the long run of many generations it’s easy to see improvement, but it’s rare for humans to display that level of foresight. Mueller et al. think a key factor is plasticity. Plasticity means that one genotype can produce multiple phenotypes. In the case of erect knotweed, they not that it does much better when there are fewer competing plants around it, producing more fruit. In this case, some work produces a measurable difference in the short term. The long term effects of domestication are an unforeseen consequence of short term advantage.

I think it’s a good paper. What makes it so is they as the authors write, they also take into account what they expect the reader to know – even if what the reader knows is wrong. For example, they note that maize wasn’t adopted in eastern North America till around 900CE. This surprised me, as when I did my MPhil in Archaeology I was sure that the date given was around 1CE. In the paper is a note, for people like me, saying that it was thought maize was established earlier, but new data favours a later date. They’d be just as right if they’d baldly stated that maize was a late arrival in the region, but by making the effort to explain why a reader might be confused they are a lot more convincing. I’d expect them to be better read on the subject of American agriculture than me anyway, but this is a helpful demonstration and points me to where I can read more if I want to get up to speed.

Another helpful point is they conclude by saying why the research matters. In this case, some of the plants are threatened by invasive species. If we’re interested in food security, then preserving this plants and re-domesticating them would provide new options for food supply. If you’re looking for the Next Big Thing, they find that goosefoot offers yields of 1,152lbs per acre (1300 kg/ha) of edible seed, compared to 450–1,070lbs per acre (500-1200 kg/ha) of quinoa using similar methods.

It might be that, from a distance of five thousand years, native Americans will be teaching everyone how to farm once again.