Why Do Trees Drop So Many Seeds One Year, and Then Hardly Any the Next?

A mighty oak covers the ground in piles of acorns. Squirrels gather them up, growing fat on the rich bounty and storing more of the seeds away for the winter.

If you live in the temperate Northern Hemisphere, this may be one of your most familiar natural scenes of autumn, because various species of both oak trees and squirrels are common and visible throughout that range. It represents a simple predator-prey model that ecologists have studied since the early days of their science. But those early scientists—and, long before them, Indigenous people—noticed something unusual about this scene. Some years, a glut of acorns fell, creating a squirrel’s paradise under every tree. Other years, almost none did.

Plants dropping most of their seeds together in one year, then taking years almost or completely off from seed production, is called “seed masting.” Oak trees are one example, but thousands of species of trees and other long-lived plants use this boom-and-bust strategy. The most common explanation has involved those hungry squirrels, birds and countless other species that eat acorns. Drop enough seeds at once, the theory says, and some will survive the predators’ feast. Ecologists call this the “predator satiation hypothesis,” and it has been a widely accepted explanation for seed masting for decades.

But predator satiation is far from the only theory. Another idea suggests masting helps insects like bees most efficiently pollinate a plant. If all of the trees of one species flower and set seed at once, that theory goes, bees or other pollinators have better odds of bringing pollen directly from one tree to another. But what if something less obvious and visible than either of these theories helped explain this phenomenon? What if the force driving it was something much smaller than a squirrel, or even a bee?

Researchers in Canada published a paper this past February in Current Biology proposing a new hypothesis for the evolution of seed masting: disease. While acorns are being gobbled up from above by hungry squirrels, they are also being attacked from below, and within, by fungi, bacteria and other pathogens. Scientists have understood for a long time that these agents can kill large numbers of seeds, but their role in determining the timing of seed release has been largely ignored. But some scientists wondered whether masting trees could drop fewer seeds in some years to break cycles of disease, rather than just to overwhelm predators in high years.

“Look at what farmers do,” says Jonathan Davies, a botanist and forest conservation scientist at the University of British Columbia, and one of the authors of the recent paper. “They often let the fields lie fallow, and that clears the pests and pathogens. You remove the crop for two, three or four years. It clears pathogens and pests from that field, and you can plant again.”

The idea that disease could play an important role was born, as many ideas are, not in a formal lab but in a casual conversation. Davies was talking to plant community ecologist Janneke Hille Ris Lambers of ETH Zurich about how variable the seed production was on the trees she studied in Washington state. The concept of pathogens as a driver came up, and Davies assumed that someone would have looked at that possibility before. But when he searched for references in journal databases, he was surprised to find an empty results screen.

“There was literally nothing in the literature about it,” says Davies.

Collecting data to support a theory like this would take decades, because of the time scales that govern tree reproduction. But before the pathogen escape hypothesis could be tested in the field, a solid foundation would have to be built, to make sure it worked even in theory. To start that process, the paper’s other author, math professor Ailene MacPherson of Simon Fraser University, came in and did what mathematical biologists do: She built a model.

The basic units of ecological theory are models, simplified representations of natural relationships that are expressed using math. Ecological models can be extremely complex, accounting for multiple species, environmental conditions and other variables. Since they were starting from a clean slate in terms of past research on the subject, MacPherson chose to use mathematical models that were as simple as possible.

“The idea was not to build the most robust models ever,” says MacPherson about the paper’s math, which she sees as a starting point and hopefully a launchpad for other researchers. “Our models are very much focused on illustrating that there might be a reason to study this.”

The closest thing to a pathogen model for seed masting in the literature was a 1992 study that looked at parasites. The paper used a version of a standard predator-prey model, like the squirrel and acorn, with basically two moving parts: seed and parasite. To adapt it for the new hypothesis, MacPherson considered two different ways that pathogens can spread: direct and environmental. Direct transmission spreads from one host to another. Environmental transmission can involve another step, either an intermediate host or another sort of reservoir where a pathogen can live between infections. A classic example is the bacterium that causes plague, which can be carried by rodents and then transmitted to humans through fleas.

Whichever method the pathogen uses, direct or environmental, there are two kinds of hosts to consider in a model: the already infected, and the susceptible, or not yet infected. According to MacPherson’s models, seed masting creates many susceptible seeds at once. In slow seeding years, the number of susceptible seeds can be so low that it could starve the next epidemic of hosts, cutting it off before it begins.

Now that the first steps of the theory are in place, Davies and MacPherson hope that other researchers can take the next steps, using more complex models and testing the theory against data in the field. One scientist who might incorporate some aspects of the theory into her work is the ecologist whose conversation with Davies sparked the idea in the first place.

Hille Ris Lambers has been studying trees and their population dynamics in Mount Rainier National Park since 2007. That data set has only recently gotten long enough, 16 years and counting, to start looking at masting patterns and, potentially, their relationship with disease. She finds the recent paper a promising start.

“I thought it was really nicely written and convincing to me that, yes, this is something that we’ve ignored as a potential long-term driver of some of these dynamics,” she says.

Rather than unseating predator satiation or pollinator efficiency as a leading theory, pathogen escape may just add to a mixture of drivers that all work together to push plant species toward masting.

“The reality is, there’s probably no one explanation,” says Davies. “This is probably going to be part of the explanation when we put this puzzle together.”

Ian Rose | READ MORE