In February 1953, working at the University of Cambridge, James Watson and Francis Crick built a physical model of DNA's molecular structure that solved a problem biology had been circling for years: how a molecule could carry hereditary information and copy itself accurately from one generation of cells to the next. Their model showed two long strands, each a chain of alternating sugar and phosphate groups, wound around each other in a double helix, with four kinds of nitrogen-containing bases, adenine, thymine, guanine, and cytosine, projecting inward from each strand like the rungs of a twisted ladder.
The structure's real insight was in how the bases pair up. Adenine on one strand always pairs with thymine on the other, and guanine always pairs with cytosine, a rule the biochemist Erwin Chargaff had already noticed in the relative quantities of the four bases without knowing why it held. Watson and Crick recognized that this fixed pairing meant each strand contained enough information to reconstruct its partner, so that if the two strands separated, each could serve as a template for building a new complementary strand, an elegant, self-evident mechanism for how a cell copies its genetic material every time it divides.
Watson and Crick published their model in a short paper in Nature in April 1953, closing with a now-famous line of understatement noting that the pairing they had found suggested a possible copying mechanism for genetic material. The semi-conservative nature of that copying, each new DNA molecule retaining one original strand and one newly built one, was confirmed experimentally five years later by Matthew Meselson and Franklin Stahl, in an experiment often described as one of the most beautiful in the history of biology.