On this day in 1992, Barbara McClintock died at ninety. She had spent most of her working life with maize — planting it, pollinating it by hand, harvesting it, and then sitting down with the cobs to count.

Here is the mechanism, in kitchen terms. Think of a gene for colour in a kernel as a tap. Something can sit on the pipe and block it, and if that blocking piece moves away during the plant's development, the tap opens again and that patch of kernel makes pigment. What you see on the cob is speckling: spots and streaks, some kernels mottled, some plain. McClintock's insight was that the pattern was not noise. It was a record of pieces of chromosome jumping from one place to another, at particular times, in particular cells. She could read the timing of the jump from the size of the spot — early move, big patch; late move, a fleck.

She worked that out in the 1940s, largely on her own, from cobs she had grown herself, and had to hold it for years while the field caught up. Transposable elements are now ordinary textbook material, and the Nobel came in 1983. What I admire is not the vindication but the bookkeeping: decade after decade of careful notes on which kernel came out speckled, until the corn finally said something nobody could argue with.