Saramsha

No. 033  ·  19 October 2026

The Slider Holds Nothing Shut

The metal slider on a zip looks like the part doing the work, but it never grips the fabric at all — it just drags a wedge-shaped channel over two rows of identical hooks, camming them into each other at exactly the angle their own shape needs to lock. Drag the slider, then bend one hook and try to close past it.

Inside the slider: a single wedge-shaped guide. The tapered side cams the two rows of hooks together; the blunt side pries them apart.

About this piece

A hook, not a zip

Nothing about a zip fastener actually zips. The word is a brand name — B.F. Goodrich coined it in 1923 for a rubber boot that closed with one of these, and it stuck to the mechanism itself. The real invention came three decades earlier: Whitcomb Judson patented a clasp-locker in 1893 that was clumsy and prone to springing open, and it took a Swedish-American engineer, Gideon Sundback, working at the same company, to redesign it in 1913 into something close to what is in your pocket right now — identical interlocking teeth stamped or moulded directly onto two cloth tapes, with a single slider to engage or disengage the whole row in one pull.

Each tooth is shaped with a small rounded ball on its front face and a matching socket hollowed into its back. The two tapes are built with their teeth offset by exactly half a step, so a ball from the left row sits opposite a socket from the right row, and vice versa all the way down. Press the two tapes flat against each other and every ball on one side seats into the socket facing it on the other — a long chain of identical ball-and-socket joints that resists being pulled apart sideways (which is the direction the fabric is pulling it, every time you wear the thing) while still being perfectly happy to bend along its length, which is why a zip can run down a curved seam.

What the slider is actually for

The slider supplies no clamping force of its own. Cut one open and the mechanism inside is a single wedge-shaped guide sitting in a Y-shaped channel — two channels at the back, for the two separate rows of teeth, narrowing into one channel at the front. Pull the slider in one direction and the taper of that Y presses the two rows of teeth together at the only angle their balls and sockets can seat; pull it the other way and the same taper levers them apart, tooth by tooth, the instant each one reaches the channel's blunt side. The diagram above the controls is that channel, straightened out and enlarged: the hooks do the holding, the wedge only ever presents them at the correct angle, and it does that identically whichever way you are pulling.

Why one bent hook stops the whole thing

Seating a ball into a socket needs the two to meet nearly square-on — bend a single tooth even a little out of its plane and its ball arrives at an angle the facing socket was never shaped to take. The wedge keeps pushing, because it has no way to know one tooth is wrong, and the result is a jam: the slider simply will not travel past that tooth in the closing direction, no matter how good the eleven, or the fifty, teeth above it are. Prising teeth apart asks much less precision of the wedge, which is why a zip with a damaged tooth will usually still open past the damage even though it refuses to close past it — the asymmetry modelled in the toggle above. It is also the practical reason a jammed zip is so often unfixable by force: forcing the slider past a bent tooth tends to bend the next one as well, and the one after that, each failure manufacturing the next.

The zip above is drawn schematically — eighteen tooth positions, a slider with no mass and no friction, and a "bend" that always lands a few teeth below wherever the slider currently sits so there is open zip left to test it against. Real tooth counts, pitches and ball/socket tolerances vary by manufacturer and by whether the teeth are metal, moulded plastic, or coil.