They gathered in secret, in the dead of night. To find their way, they consulted a map originally drawn more than a century earlier, counting paces and triangulating their position with compasses. When they reached what they hoped was the right spot, they began to dig.

It sounds like something from a movie, but it’s a scene that has played out over and over in real life — and will again in about 17 years. The treasure-hunters are a group of scientists at Michigan State University, taking part in one of the world’s longest-running experiments. Instead of a chest of gold, their treasure is a vault of seeds buried in 1879. Their ongoing mission, passed down through the years, is to answer a seemingly simple question: How long do seeds retain the ability to grow?

Technically speaking, a seed is a plant embryo wrapped, along with a reserve of food, in a protective coating. It’s also a tiny bundle of potential: a blade of grass, a fragrant flower, a towering tree, neatly packaged and waiting until the conditions are right to grow. But seeds can’t wait around forever — after a while, they lose the ability to sprout.

“They’re strange because their job is to stay exactly the same, until they transform. It makes it really tricky to study them, because the best test we have for whether or not a seed can germinate is by trying to germinate it,” says Margaret (Grace) Fleming, an assistant professor at MSU’s Department of Plant, Soil and Microbial Sciences.

Varied Results

The question of seed longevity interested Professor William J. Beal, who was hired at Michigan State University in 1870. MSU, or as it was then known, the State Agricultural College, had an extension program called the Farmer’s Institute, in which scientists from the college would meet with farmers to share the latest news in agricultural science and answer the farmers’ questions.

“One of the big questions in agriculture, that’s always been an issue, is weeds,” says Frank Telewski, professor and director emeritus of the W.J. Beal Botanical Garden and Campus Arboretum at MSU. “I can see where farmers would be asking Professor Beal, ‘How long do we have to keep weeding our fields in order to get rid of these pesky weeds and increase our crop efficiency?’” Essentially, they wanted to know how long weed seeds already in the ground would continue to wreak havoc on new crops.

To answer this question, Beal devised an experiment. In the fall of 1879, he gathered 20 glass bottles and filled each of them with sandy Michigan soil containing 50 seeds of 23 common weeds. He then buried them 20 inches deep, with the bottles open and slanted downward to allow for gas exchange but prevent water from collecting. In doing so, he isolated them while replicating the conditions experienced by seeds buried underground. Then, he waited.

In 1884, he dug up the first bottle and attempted to sprout the seeds. The results varied wildly from species to species — of the 50 shepherd’s purse seeds, all 50 germinated, as did 45 curled dock, 38 evening primrose, 26 stinking chamomile, and three water pepper. On the other hand, none of the ragweed, black mustard, spotted spurge, arborvitae, corncockle, rye brome, American burnweed, or broadleaf plantain seeds sprouted.

Beal continued his experiment for the next three decades, digging up a bottle every five years and documenting which seeds grew. When he retired in 1910, he handed the project off to a colleague, Henry Townsend Darlington. Darlington noticed that the same seeds tended to sprout at every five-year interval, and saw an opportunity.

“He said, ‘This is really great, but instead of every five years, why don’t we move it to every 10 years, and we’ll stretch this experiment out and see what happens,’” says Telewski. As the project continued to move along to new generations of scientists, they continued to see the same usual suspects germinating; in 1980, with six bottles then remaining, the MSU faculty voted to extend the interval to one bottle every 20 years. The experiment is set to continue through 2100.

Hidden and Undisturbed

Telewski remembers reading news stories about the 1980 bottle being unearthed. “I was just a lowly graduate student” starting a Ph.D. in botany, Telewski says. He recalls thinking, “This is really neat, what a great experiment. But I never even thought or even dreamt that I would be involved.”

By the time the year 2000 rolled around, though, Telewski had taken a position at MSU, and one of his senior colleagues, Jan Zeevaart, invited him to join the experiment. “I was just gobsmacked,” recalls Telewski. “On a fine spring day in 2000, Jan and I met on campus and we excavated a bottle.” They were able to generate 23 sprouts of Verbascum, a tall spiky plant with pale flowers (also known by the common name mullein), as well as one Malva rotundifolia, a mallow plant with rounded leaves.

In 2017, Telewski began recruiting a younger generation of scientists to take up the experiment for him after he retired. His timing, it turned out, underlined the importance of multiple collaborators on the project — just months after recruiting one of his new partners, David Lowry, Telewski had a stroke. As he recovered, he and Lowry discussed their hopes for the project.

“We started talking about, how do we bring the Beal experiment into the 21st century?” says Telewski. “What kinds of questions can we ask, what technologies do we have that we didn’t have before? What areas of science are going to be really interested in this besides just the simple question of longevity and weedy species?”

With these questions in mind, they recruited three more partners: Lars Brudvig, Marjorie Weber, and Margaret Fleming. These scientists brought with them research experience in restoring landscapes to their original flora and fauna, examining how ecological relationships influence evolution, and studying the molecular biology behind seeds’ metabolic activities.

The new recruits’ first encounter with the Beal seeds was delayed a year by COVID-19, but on a snowy night in April of 2021, they set out to retrieve the fifth-to-last bottle. The mission took place in the wee hours, both to prevent the seeds from getting exposed to sunlight that could trigger germination, and to keep the bottles’ location hidden and undisturbed. “We all arrived at our meeting spot, and we were feeling so clandestine about it. It was like being in a secret society,” says Fleming.

In keeping with the then-142-year-old experiment, the means of finding the bottles was charmingly old-school. “It’s basically a treasure map,” says Fleming. “We had to triangulate where the hole would be based on counting paces from landmarks and the intersection of lines.” After half an hour of searching in the cold and dark, they began digging, and before long, their shovels hit the walls of the concrete vault containing Beal’s bottles.

Back in the lab, they emptied out the bottle and placed the seeds in climate-controlled growth chambers, in beds of sterile soil to ensure that no new contaminants were sprouting instead of the 142-year-old Beal seeds. A week later, right on schedule, the first baby Verbascum plants poked through. Overall, 20 Verbascum seeds sprouted.

It’s not clear why Verbascum seeds are so long-lived; the answer may lie in their permeable seed coats that allow in a small amount of water, but Fleming says, “We’ve truly no idea.” It’s one of many questions that she and her colleagues hope to answer in the coming decades of research.

Long-Term Data

To help unravel the mysteries of seed longevity, the researchers are turning to a variety of tools and disciplines that Beal never even dreamt of, including analyses of the plants’ genomes and proteins. “We’re going to do Beal 2.0, which is just replicating Beal’s experiment just as he set it up — but this time we have some of idea of when certain species are going to die, and we could do sampling around those times and look for what is deteriorating, what kinds of things are changing. Are the proteins more messed up, or is the RNA? Is it something else?” says Fleming.

The team has also set up a series of community science projects to gather more data points about seed behavior in different places besides East Lansing. “I figured that turning to the community would be more expedient than me going around and burying seeds at different places. And this way, we can get, hopefully, worldwide participation,” says Fleming. The version of the experiment adapted for non-scientists doesn’t rely on a plot of land to lie undisturbed for centuries, though — people can send in their observations after just a few months. Fleming also hopes that to get long-term data, schools could get involved and treat the seed project like a time capsule. (To sign up, visit https://citsci.org/projects/seed-burial-experiment.)

As for the original Beal experiment, four bottles remain underground, waiting to be studied. Telewski has retired since unearthing the 2021 bottle, but he hopes to join the 2040 dig as a spectator. “You’ve got this experiment that has been so dutifully taken care of and stewarded by all these previous scientists, and the baton has been passed to you to run this leg,” he says. “It’s really fascinating, and it’s really rewarding.”

This story was originally published in our May June 2024 issue. Click here to subscribe to read more stories like this one.