Only 35 Mount Graham squirrels remain in the wild, but five captive squirrels could hold the key to their long-term survival — if we can get them to breed.

Mount Graham red squirrel

It began with a bolt of lightning on June 7 and ended with a fire that eventually encompassed a staggering 48,000 acres of southeastern Arizona. By the time the blaze had been extinguished this past July, thousands of trees had been lost or damaged, impacting the already degraded habitat for the critically endangered Mount Graham squirrel (Tamiasciurus hudsonicus grahamensis). Surveys conducted this past September in the high-elevation forests of the Pinaleño Mountains, about three hours east of Phoenix, revealed that the squirrels’ population had fallen to an estimated 35 animals and that at least 80 percent of their habitat had been damaged by the fires.

Could this be the end of the Mount Graham squirrel, which was already once thought to be extinct and has been protected by the Endangered Species Act since 1987?

The answer to that question may lie not on the mountain itself but in the halls of Phoenix Zoo’s Arizona Center for Nature Conservation, where five Mount Graham squirrels form the core of a captive-assurance program that could help save the species from extinction.

There’s just one catch: We need to figure out how to get them to breed first.

That hasn’t been easy, says Stuart Wells, the zoo’s former director of conservation and science, who was in charge of the program until last month. The squirrels, it turns out, are extremely territorial, aggressive loners who attack and even kill other squirrels, including potential mates, that invade their home turf. That makes it impossible to keep the captive animals together in the same enclosure — or even within sight of each other. On top of that minor complication, the animals are also incredibly sensitive to environmental changes like temperature and sound. And until recently we simply didn’t know how to keep the species healthy in captivity, let alone get it to breed.

Fortunately we’ve learned a lot since the squirrels were first brought into captivity in 2011. Wells says one of the most striking new pieces of information we’ve discovered is that female squirrels don’t only enter estrus once or twice a year, as most previous scientific evidence indicated. Instead, it appears they cycle about every 25 days.

Wells and his team hit upon this new information in the early days of the captive program, before they had a federal permit to actually breed the animals (a required step under the Endangered Species Act). As part of ongoing health monitoring they tested the female squirrels’ droppings for steroids called fecal metabolites — a technique Wells first used on cheetahs — which revealed when the animals were fertile. This is information that never could have been gathered in the wild, he points out.

“We were actually very surprised when we got the first year’s results back and noticed that they were actually cycling somewhat periodically throughout the year,” Wells says. That meant they had more than one opportunity a year to try breeding.

After timing, the next challenge was figuring out which of the zoo’s three males would be welcome suitors to the two females in breeding season. Again, not an easy task, since the males and females had to be kept apart most of the time in order to minimize their aggression toward each other. Wells and his team solved that problem with more steroid tests, which revealed that males became much less aggressive when females were most receptive to breeding.

That was enough information to try to put a pair together. On the zoo’s first try, in 2016, they got a successful breeding attempt, and the female became pregnant.

It didn’t come to term, though. Tragically, the zoo’s air conditioning went out and temperatures in the squirrels’ enclosures soared above comfortable levels. “It only got to 82 degrees in the enclosures, but these are animals that live at 10,000 feet,” Wells says. With temperatures above what the squirrels would normally encounter high in the mountains, the pregnancy failed. Another test of her fecal steroids revealed why: the female’s stress cortisol levels had shot up to 10,000 nanograms, well above her normal level of just 488, because of the heat.

A second attempt also failed because of a different stressor: noise. “Our breeding season last year began in March and concluded in October,” Wells says. That overlapped with the time the zoo was building a new enclosure to hold the squirrels. “It wasn’t heavy construction,” he says, but it was too loud for the animals. “If you can imagine how much sound you hear in the forest when you’re walking through, that’s pretty much where they’ve evolved. They tend to avoid any sounds above 70 decibels, but in this captive setting their tendency to want to move away from that sound would be compromised because they can’t go someplace else.”

That stress was too much: The animals just weren’t in the mood.

Those early attempts didn’t work, but they helped improve knowledge of what will be necessary in the future to allow the rare squirrels to breed. “What we’re hoping is that this next coming season will have everything in place and be ready to get a successful breeding, and that’s going to be the next part of the story,” Wells says.

And if they do succeed in breeding, it could be a game-changer for Mount Graham squirrel conservation. The captive females could conceivably give birth a few times a year, each time producing two to four pups, some or all of which could eventually be returned to their native habitat. “The goal of the program is producing animals that can survive in the wild,” Wells says.

Here’s another interesting twist: The research conducted to benefit the captive population might also be of value for the few remaining wild squirrels — not in terms of getting them to breed, but of making sure they have enough food on the mountaintop, which was heavily degraded by construction even before the fires. “Some of the work we did early on was to develop a nutrient program for keeping these guys at the right weight in captivity without being too heavy,” Wells says. That information could be useful for providing supplemental food for wild squirrels whose seed sources were lost or damaged in the fires. “Our partners with U.S. Fish and Wildlife and Arizona Fish and Game have identified in which areas the squirrels are active. Providing food to those areas will help the squirrels have as good a chance as possible of having food for the winter.”

And not just this winter — Wells says it could take up to 70 years for the trees on Mount Graham to recover enough to provide enough food for the squirrels and protective cover from aerial predators. “That is…daunting,” he says. “How do you keep these guys going for 70 years, and none of us will be around — at least I won’t — to say ‘yeah, that worked.’ You kind of have to have faith that what you’re doing now is actually going to have a positive impact. Really, that’s all you can do.”

John R. Platt

is the editor of The Revelator. An award-winning environmental journalist, his work has appeared in Scientific American, Audubon, Motherboard, and numerous other magazines and publications. His “Extinction Countdown” column has run continuously since 2004 and has covered news and science related to more than 1,000 endangered species. He is a member of the Society of Environmental Journalists and the National Association of Science Writers. John lives on the outskirts of Portland, Ore., where he finds himself surrounded by animals and cartoonists.