Some monarch butterfly populations don’t migrate. They live in tropical areas of the world, and don’t get as much press as the North American populations that show up en masse on weather radars. But for scientists who study monarchs, they’re critical to understanding the bugs. They’re the reason that most researchers think that these insects first evolved in South America as a non-migratory species, and only started migrating a few hundred years ago, once they moved into the US. And, if a study published today in Nature is further supported, they’ll also be the reason that we were just plain wrong.

Most people — including scientists — think of migration when they think of monarchs, said Marcus Kronforst, an evolutionary biologist at the University of Chicago and a co-author of the study. That may be why so many biologists think that this behavior is one that the butterflies evolved into having, rather than one that came standard and which some monarchs subsequently lost. But today's study suggests that migration predates stationary monarchs.

By taking genetic data from over 100 butterflies and mapping it out, the researchers were able to establish that not only did the monarch start out as a migratory species, but that the non-migratory populations that live in Europe, South America, and Hawaii are actually the product of independent dispersal events.

"It looks like the most closely related butterfly to the monarch itself — the southern monarch — is also migratory," Kronforst says. Which means that at some point, millions of years ago, a butterfly evolved migration, and then split into two separate species. This, Kronforst says, is how the migratory North American Monarch came to be.

But that was far from the researchers’ most surprising finding. Because, as it turns out, only a tiny part of the butterfly's genome substantially contributes to monarch migration. And every time the researchers ran their analyses, a single gene popped up as being wildly different when comparing migratory and nonmigratory populations — a gene that’s actually related to flight muscle function. It participates in muscle metabolism, so when it’s expressed at high levels, it helps butterflies fly more powerfully.

"So clearly, we thought the migratory monarchs would have more of this gene," he says. But when they put monarchs in flight chambers, the researchers realized that this simply wasn’t the case. Migratory monarchs don't express the gene as much, so when they travel long distances, they consume less oxygen, and are therefore more efficient. "And they don't show that difference when they aren't flying," Kronforst says, "so we know it’s related to this gene."

If he has to guess, Kronforst says, his bet would be that butterflies who don’t migrate probably need to fly harder and faster so they can chase mates or evade predators. Migratory monarchs don’t experience the intense competitive pressures that non-migratory butterflies do, Kronforst says, so they can get away with having slower metabolisms and flying more slowly.

"The section on demographic history is spectacular," says Francis Villablanca, an evolutionary biologist at California Polytechnic State University who did not participate in the study. The paper, he says, provides answers to many questions that researchers have long wished someone could answer. "In one fell swoop, this paper provides the context for us to be able to think about and understand the origin of and the differences between populations."

The study’s findings also matter for monarch conservation, Villablanca says. Monarch populations have been dwindling because humans have waged a war against the monarch’s host plant, the milkweed. But this study shows that each monarch population is genetically unique, which means protecting as many monarch populations as we can suddenly becomes more important. "They may all be descendants of an ancestral North American population," he says, "but it would be difficult to recreate any of them because now they contain unique characteristics."

Moreover, Kronforst says, most scientists think that monarch migration is just a few hundred years old. But his study’s results suggest that the behavior is much older than that — thousands upon thousands of years older. So "if you think of migration as just being a very recently evolved thing, then maybe you care less if it disappears," he says. But "it’s actually very old, and we might be witnessing the very last days of it." For some reason, he says, that seems far worse.