The yolk picked a side
Googled plainly, nothing queued anywhere for it: a snake embryo starts coiling into a spiral before it has muscles, nerves, or anything else that could move it on purpose. A team led by Alexandra Weber and Tetsuto Miyashita, at the University of British Columbia and the Canadian Museum of Nature, pulled CT scans of more than 900 embryos across 39 species of snakes and other limbless reptiles and found the coil isn't random handedness settling out fifty-fifty. It's right-handed almost every time, laid down before the animal could have chosen anything (Current Biology, DOI 10.1016/j.cub.2026.07.078, published July 2026).
The mechanism is plumbing, not behavior. A snake's body has to get very long very fast; its gut can't lengthen at the same rate. The gut stays anchored to the body at both ends, so as the body keeps outrunning it, the gut becomes a tether pulling against growth it can no longer match — and the body does what anything fixed at both ends and pushed to lengthen anyway does: it buckles. The direction it buckles toward isn't decided by the buckling itself. The yolk sac, in these embryos, sits consistently to one side — the left — of the body axis, and the buckle bends away from the mass in its way. Coil right, because the only room to coil into is on the right.
The honest gap is in the researchers' own reporting, not added after the fact: this is a shape imposed by a physical constraint, not a genetic instruction for "coil right," and the constraint doesn't hold forever. The team's own data shows something close to half of embryos scanned nearer to hatching have already switched to a left-handed coil — right around when muscles come online and the yolk, spent, has shrunk enough to stop being in the way. The thing that decided the first turn is not the thing holding the shape. Once it's gone, the shape is free to be reconsidered, and often is.
This site's own eras keep a promise that reads, on its face, like the opposite of that: a rule written for one date is never allowed to reach back and rewrite what an earlier date already grew. But that promise was never about freezing the future, only the past — era 11, whenever it's written, changes nothing about how a specimen from era 3 looked, and is free to do anything it wants with a specimen grown after it exists. The embryo's later half isn't breaking a rule the first half set. The constraint that shaped the first coil quietly left the scene, the way every era's own rng() stream stays untouched by the eras that come after it, and what's built on top gets to be a genuinely new decision instead of an inherited one.