Charles Platiau / Reuters

In April, researchers at Tufts University posed a nutrition riddle. They compared people who took vitamin pills with people who got the same nutrients the old-fashioned way, by eating food.

Tracking intake of vitamins A and K, magnesium, and zinc, the scientists found that people were less likely to die of heart attacks and other diseases when these nutrients occurred in their diets. As the Tufts researcher Fang Fang Zhang said at the time, “There are beneficial associations with nutrients from foods that aren’t seen with supplements.”

Many vitamin supplements are synthesized to be exact replicas of the compounds you’d get from eating an apple or an orange. The chemistry should have the same effects on the body. Unless, of course, something was missing from the equation.

In a similar puzzle, recent studies have illuminated harms associated with highly processed foods—even though many of these foods are packed with added vitamins. White pastas and breakfast cereals, for example, may contain an entire day’s worth of some vitamins (synthesized and added, sometimes by law). As long as we’re getting the nutrients, why should it matter whether food is “processed”? Is processing simply bad?

One explanation for the benefits of eating minimally processed foods is probably fiber, which processing often strips away. Fiber slows the absorption of sugars, so they don’t hit our blood as quickly and cause insulin to spike (as with eating an apple versus drinking apple juice). Fiber also feeds our microbes. People with low-fiber diets have less diverse gut microbes—the trillions of microorganisms that populate our bowels and are vital to our digestion, metabolic health, and the functioning of our immune systems. The best known indicator of a healthy biome is diversity.

But fresh produce and grains also give us more than fiber. An exciting, emerging idea is that fruits and vegetables are healthier than the sum of their parts, not just because of nutrients and fibrous skeletons, but because they contain microbes themselves.

That might seem like a bad thing. But it actually builds on a story I wrote last week about how the immune system, gut microbes, and the food we eat all work in harmony to influence weight gain and loss. The closest thing to practical advice from scientists was to maintain a “diverse biome.” But how do people actually do that? Many readers wrote to ask for more concrete advice. (“Sounds like you still want us to take probiotics every day?”; “What’s the best probiotic?”; “Can I buy your microbiome?”)

Doctors have insisted for decades that unnecessary antibiotics should be avoided, to prevent the evolution of antibiotic-resistant superbugs. Upsetting one’s own personal microbial diversity adds yet another reason. Fermented foods of course contain bacteria, and their consumption has been linked to some health benefits. Beyond that, many people believe it’s necessary to turn to supplements. Even Harvard Medical School’s website tells patients as much, advising that “there are two ways to get more good bacteria into your gut: fermented foods and dietary supplements.”

But supplements are an enormous and barely regulated industry. Even the best clinical trials are limited and short-term. Taking a probiotic supplement of Akkermansia was found last month to have some metabolic benefits—but the same bacteria are also associated with multiple sclerosis. Such things are not to be wantonly introduced into everyone’s guts, but used strategically in specific populations with specific needs—more like a drug than a food.

For all of human history, the gut microbiome has gone without bacterial pills. Fermented foods have been part of many cuisines around the world, but our ancestors didn’t live on kombucha. There had to be another source.

And, it turns out, there is: fresh produce.

In a study from July in Frontiers in Microbiology, researchers found that the average apple contains about 100 million bacteria. Most are inside, not on the skin. They came from many different taxa—as opposed to the probiotic-supplement pills, which tend to be only one type of bacteria. Of the millions of bacteria in any given apple, very rarely are any the sort that cause diseases; most are innocuous or even beneficial.

The idea, the apple researchers explain, is that these bacteria join and interact with the trillions of microbes that are in our guts already—which are vital to our digestion and metabolic health, and the functioning of our immune systems. Food is the main way that our gut biomes are populated throughout our lives, and microbe-rich foods seem to be important to maintaining diversity. The researchers suggest that microbial profiles could eventually become standard information on nutrition labels (currently limited to fats, proteins, carbohydrates, vitamins, and minerals).

When it comes to apples, most of the microbes turn out to be in the core, central part, which most people don’t eat, because it is fibrous—full of fiber and microbes. If you eat only the flesh and skin, you miss out on 90 percent of the bacteria, some of which are the same species sold in expensive pills at Whole Foods. As I’ve argued in the past, if you eat the apple from bottom to top, the fibrous “core” is barely noticeable. The seeds of the apples had the most microbes of any part. They do contain trace amounts of cyanide, but adults should have no problem with a single daily core.

At Rutgers University, Donald Schaffner, a food-science professor, does not eat apple cores. But he is intrigued by the idea—and by the apple bacterial counts. His team has been counting microbes in food for years. Its main concern has been looking for disease-causing bacteria. The diversity of microbes in an apple comes as news even to him—and the numbers would have seemed impossible to him not long ago.

“This microbiome research is blowing things wide open in terms of complexity,” he told me. When the Rutgers lab started studying foods, the only way to look for microbes was to culture bacteria. It turns out that this was detecting only a small percentage of microbes, because not all of them grow on agar. Newer technology allows scientists to test for DNA, and this has revealed orders of magnitude more microbes on and in our food than previously imagined.

“We’ve known for a long time that there are organisms in fermented foods that have benefits,” said Schaffner, who had just eaten yogurt, “but there is a lot more to it than that.”

Each week his research team samples foods in the dining hall at Rutgers. Team members bring, for example, an egg back to the lab and mix it in with some dilution and put it into a “stomacher,” a sort of glorified sack that churns and shakes to simulate the action of the food being partly digested in the stomach. Then the team tests the slurry for bacteria—what’s there that would make it through the acidic barrier of the stomach. Produce consistently has more organisms than other foods.

“As long as it’s not spoiled, that may not be a bad thing,” Schaffner said. “You always want to limit human pathogens, but you also want to look at the overall microbiota.”

If the stomach-machine test found lots of microbes in a salad, that would be expected; it would only be a problem if a disease-causing species like E. coli appeared. By contrast, even a small number of bacteria on a hard-boiled egg suggests something is awry. “One of the foods I passed up on the breakfast buffet this morning is hard-boiled eggs,” he said. These should be relatively microbe-free, but this is often not the case. They represent “an excellent environment for growing bacteria.”

Short of food poisoning, the idea that foods with naturally higher bacterial counts could be good for human health is promising. It also offers a plausible explanation for why what we already knew to be true is indeed true. If fresh produce can be considered a probiotic food, that would only be cause to double down on the old nutritional wisdom: Eat a “balanced” diet, full of fresh fruits and vegetables, nuts and seeds, et cetera. If you do all that, except for specific cases, the average person shouldn’t need supplemental microbes.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.