Wood frogs live across northern North America, up into Alaska, and they get through winter by shutting down further than sounds survivable. The heart does not slow for the cold months. It stops. Breathing stops, circulation stops, and no measurable brain activity remains. In spring, the same heart starts again on its own.
The frog, Rana sylvatica, also listed as Lithobates sylvaticus, does this under the leaf litter on the forest floor, not sealed inside a block of pond ice, which is how the story usually gets illustrated. While it sits there, a large fraction of its body water turns to ice. It is rigid, there is frost on its skin, and by most of the checks a person would think to run, it looks dead. The accurate version is narrower: the heart has stopped and the brain shows nothing measurable, and in spring the animal resumes normal activity, which settles the question.
Where the ice actually forms
The word frozen is doing careful work here. In 1994, Rubinsky and colleagues put freeze-tolerant frogs in an MRI and imaged the freezing and thawing directly. The ice forms predominantly in the extracellular compartments, the spaces between the cells and the body cavity, while the cells themselves stay unfrozen. The tempting version, that ice never gets inside the cells, overstates what the imaging shows. So does calling the animal frozen solid. A wood frog in midwinter is closer to a container of unfrozen cells packed in ice.
The protection is sugar. As freezing begins, the frog's liver breaks down its glycogen stores and floods the bloodstream and the cells with glucose. Storey and Storey documented this in 1984: rapid glycogenolysis in the liver and large increases in tissue glucose, right as the freeze sets in.
What the glucose does
Glucose here is a cryoprotectant. It protects the cells against the freezing itself, against dehydration, and against ice damage. There is a clean experiment behind that: in 1993, Costanzo, Lee and Lortz raised tissue glucose concentrations directly and found the frogs survived low subzero temperatures better.
The verb matters, though. That result supports saying glucose helps, that it improves survival. It does not support saying glucose is why the frog lives. The sugar flood is one of several mechanisms running at once, and flattening it into the whole explanation is how this animal usually gets oversold.
Two details make it stranger. The freeze is not a single event: a wood frog can freeze and thaw repeatedly across one winter, the heart restarting on its own each time. And the trait scales with climate. Alaskan wood frogs tolerate far colder and far longer freezing than southern populations, which Costanzo and colleagues documented in a subarctic Alaskan population in 2015.
Spring, and the numbers I left out
When the thaw comes, nothing external restarts the heart. It starts on its own, circulation resumes, and the frog goes back to being a frog. You may notice this page quotes no temperatures, no freeze durations and no survival percentages. That is deliberate. The figures differ by population, the Alaska result being exactly that difference, and secondhand retellings tend to garble them, so I am leaving the numbers to the papers listed below.
Last thing, bias disclosed: I build Curio, an iOS app that serves five short curiosity stories a day, and the wood frog is exactly the kind of story that ends up in it. If you want a few of these each morning, it is Curio on the App Store. If not, the Rubinsky paper is the one to read. A frog in an MRI machine is worth ten minutes of anyone's time.