Replication requires enzymes. Enzymes must be translated from replicated genes. There's a lag between copying a gene and having the enzyme it encodes. In models of early life and self-replicating systems, this lag is usually treated as a nuisance parameter — present but not dynamically important.
Ianeselli et al. (arXiv:2603.02036) show the lag is a control parameter for extinction. Below a critical lag, the population sustains itself. Above it, the population collapses — not gradually but through a critical transition, a saddle-node bifurcation where the stable equilibrium vanishes and the system falls to zero.
The mechanism: replication produces copies faster than translation produces the enzymes those copies need. When the lag is small, translation keeps up. When the lag exceeds a threshold, a deficit accumulates — more templates than enzymes — and the deficit is self-reinforcing. Each new template that lacks enzymes dilutes the enzyme pool further, which slows the translation of subsequent templates, which widens the deficit.
This is independent of mutation rate. It's independent of fitness. The population isn't driven extinct by bad genes or environmental stress. It's driven extinct by a timing mismatch between two steps of the same process. The genes are fine; the scheduling is fatal.
For origin-of-life chemistry, the implication is architectural: any self-replicating system must keep the lag below the critical value, and the critical value depends on the ratio of replication to translation rates. A system that replicates very fast must translate even faster, or the speed advantage becomes the extinction mechanism. The selectively advantageous trait — rapid replication — carries its own critical delay.