The first fossil was so stunning that Brian Hebert nearly missed the second.

Tucked into the stump of a 300-million-year-old tree was a neat line of vertebrae, sprouting a series of delicate, wispy ribs. A smattering of belly scales freckled the space below, paving a path to a pelvis and a pair of petite thigh bones. These were the first known remains of Dendromaia unamakiensis, an early land-dwelling vertebrate that likely resembled a foot-long monitor lizard .

“I can close my eyes and remember it like it was yesterday,” says Hebert, an amateur fossil hunter who happened upon the tree in Cape Breton, Nova Scotia, in 2017. “It was three dimensional, right in front of my face.”

Then Hebert spotted another set of bones that stopped him dead in his tracks: a tiny, inch-long skull, nestled into the space where a left femur met a pubic bone. This skull, Hebert realized, belonged to a juvenile, curled up against what was probably its mother.

Hebert didn’t know it at the time, but what he found would soon become the prime piece of evidence in a paper published today in the journal Nature Ecology & Evolution arguing that parental care—investing resources in offspring after birth—is at least 306 million years old.

Plenty of today’s amniotes, the group that includes mammals, birds and reptiles, fuss over their young. While the evolutionary strategy is costly, it increases the chances that an animal’s offspring will succeed, and parenting has often been considered by researchers to be a trait of more modern animals. But this ancient fossil duo, which dates back to the era before dinosaurs when our egg-laying predecessors first crawled ashore, suggests the origins of this nurturing behavior is much more deeply rooted in this branch of the tree of life.

“We tend to think of animals in [this part of] the past as ‘primitive’ or ‘simple,’” says Jackie Lungmus , a vertebrate paleontologist at the University of Chicago who wasn’t involved in the study. “But they deserve more credit. Even back then … these animals were probably doing a lot of the things that animals still do today.”

Even before he’d left the stump, Hebert, who’s been scouring the Nova Scotian landscape for fossils since childhood, knew he had discovered something big. To confirm the find, he recruited paleontologists Hillary Maddin and Arjan Mann , who carefully transported the bones back to their laboratory at Carleton University in Ottawa for further analysis.

Under the microscope, the nature of the fossils was unmistakable, Mann says. They belonged to two individuals of the same species, one fully grown and the other young, and bore the hallmark characteristics of varanopids, an extinct family of pre-mammalian ancestors with reptilian features that lived around 300 million years ago. But the adult looked distinct enough from its relatives to earn its own genus name: Dendromaia, or “the mother in the tree.”

The fossils’ exquisite preservation hinted that the pair had died a sudden death, perhaps during a storm that flooded their stumpy hideout with suffocating sediments, preserving their final moment in freeze-frame. Swaddled between the adult’s tail and hind leg, the smaller specimen appeared as though it was being purposefully guarded from harm.

“It looked a lot like denning behavior,” says Maddin, who found it hard not to think of a protective mother cradling her baby.

Mann, her graduate student, casually quipped that they’d found “the earliest evidence of parental care.” He’d meant it as a joke, but his words reminded Maddin of a similar fossil unearthed in South Africa a decade prior: a Heleosaurus specimen that had died with its tail curled around four juveniles during the Permian, 260 million years ago. Though separated by about 45 million years, both fossils were varanopids—and both, it seemed, had died sheltering smaller versions of themselves.

The researchers who documented the Heleosaurus find had pinpointed the remains as a probable family group, giving Mann and Maddin more confidence that they’d stumbled on something similar and more ancient. Mann had hit not on a punchline, but a viable hypothesis: Since their earliest days, varanopids may have made parenting a priority. (Some researchers have previously presented even older evidence of parental care in invertebrates , but Dendromaia could represent the earliest known example of an amniote parent looking after live young.)

Without a time machine, researchers can’t know what these animals were actually doing at their times of death. Clear-cut behavioral evidence, after all, “isn’t something that gets preserved in rocks,” says Stephanie Drumheller-Horton , a vertebrate paleontologist at the University of Tennessee, Knoxville, who wasn’t involved in the study.

It remains possible, for instance, that the two Dendromaia weren’t a family unit at all, but two unacquainted refugees seeking safe harbor from a bad storm. Similar pow-wows between unrelated adults and juveniles have been seen in the fossil record before, points out Eva Hoffman , a vertebrate paleontologist at the American Museum of Natural History who wasn’t involved in the study. Even the wraparound tail doesn’t guarantee anything: Maybe the two were simply short on space. Until more examples of possible parent-child associations emerge, Hoffman says, some caution may be warranted.

But Drumheller-Horton thinks a mom hunkering down with her kid is still the most probable explanation. The placement of the fossils was also unlikely to be a fluke. Bones this delicate don’t just slosh into such intimate configurations.

And if Dendromaia and Heleosaurus were both precocious parents, “that tells us this behavior could have been present in a common ancestor of this group,” Maddin says. Perhaps paleontologists have yet to uncover the lineage’s oldest doting moms and dads.

Whatever their origins, parenting and its perks clearly stuck around. In many ways, it’s a sensible strategy, Mann says. By ferrying their kids through early life, ancient animals helped ensure their survival and the continued persistence of generations to come.

“Parental care is a strategy with a long antiquity,” he says. “Clearly, it’s worked out multiple times in evolutionary history. And we should have an appreciation for it.”