Evolution and Arithmetic

Chapter 7: Junk DNA and a Demolition Tool

← The Junk That Wasn'tLenski's Bacteria →

Point Mutations

The section above left DNA looking a great deal more interesting than the old "junk" picture allowed, full of much more than just the instructions for making proteins. But that richer picture comes with a cost. The Standard Model let evolutionary theory lean almost entirely on one simple mechanism: change one letter of DNA at a time, and let the mostly-empty genome absorb the damage while good, working genes stayed safely out of the blast radius. If large stretches of DNA are actually doing something important, that safety margin shrinks. It's time to ask how well that one simple mechanism — the single-letter change — actually holds up on its own.

Let's start with an admission. Small changes happen constantly when organisms reproduce. Every generation, in every population, copying errors creep in, one letter here, one letter there. And it isn't just that these changes happen — every now and then, one of them actually does something. A single-letter change, or a single event that affects several DNA letters at once, shows up, and an organism carrying it turns out to be better off for it: better suited to its surroundings, more likely to survive, more likely to pass that same change on to its offspring. This isn't a hypothetical. It's been documented, directly, in living populations.

Two cases make the point particularly well, and it's worth walking through both before asking what they actually prove.

Certain populations of a small fish called the stickleback have lost an entire skeletal structure — a pair of pelvic spines used for defense in the ocean, but a liability in the freshwater lakes some populations moved into. The gene involved is still there, fully intact, in the freshwater fish, and still doing its other jobs elsewhere in the body. What changed, independently, in population after population, was a nearby switch controlling where and when that gene gets used — a small stretch of regulatory DNA, not the gene itself, deleted or disrupted. The spines disappeared not because the instructions for building them were deleted, but because a switch controlling their use was turned off by a mutation.

In other words, the building of a structure is under the control of switches as much as of the proteins themselves — something any "it's all just proteins" picture leaves out.

Last revised 2026-11-19

← The Junk That Wasn'tLenski's Bacteria →