Topological Linking-Number Heredity in a Crystalline Substrate, and the Open-Ended- Evolution Wall A Negative Result, with a Mechanism Worth Keeping

We report two things, and we will tell you the ending first because the ending is half the point. The first is a specific, falsifiable mechanism by which a crystal could carry heritable information: the topological linking number of a chiral texture, Lk = Tw + Wr, written when two helical records merge and mutated in integer quanta when the crystal fractures — a topoisomerase-analog in dead matter — with the substrate's cleavage class predicting whether that inheritance is faithful or scrambled. This is, to our knowledge, new, and it is decided by a single experiment we name but did not run. The second thing we report is a failure. We built this substrate in simulation, gave it heredity and selection, and it would not cross from faithful replication into open-ended evolution under any undirected driver we tried. The population drifted toward order, not the edge; organized complexity fell. This is the wall the entire artificial-life field stands at, and we did not get over it. We map exactly where it is, name the four things that smother the spark, demonstrate the removal of three of them, and report — against our own hope — that removing them was necessary and not shown sufficient. Biology's prion, a structural template that copies faithfully and evolves almost not at all, is the confirming witness: it ran our experiment, in protein, and got our answer.