Every chain on this site ends on one of these. The house rule is become a mystery before the woo — a stated open question is stronger than an assertion that can't be defended, so nothing below gets resolved into one. If a shape is missing or wrong, that's exactly the kind of thing worth saying so.
Entropy production rate and the Lyapunov exponent are related but not identical — a bacterium in log phase has enormous entropy production and is deeply ordered; a system can sit at criticality on modest dissipation. Claiming they're one thing would need a derivation that doesn't exist yet.
Closes with: a derivation linking entropy production and the Lyapunov exponent in driven systems — contested, and a real result if it goes through. Raised in chapter 1, not yet resolved there.
The criticality literature is very good at the critical point and mostly silent on the far side. Laser threshold, Bénard rolls, superconductivity are all cases where "too much" turned out to be a new phase rather than noise — but there's no general existence condition for when a receiving structure is there to be found.
Closes with: stating the existence condition for a receiving structure — the framework's own terms would call this deciding whether an excursion is an ignition or just a mess. See chapter 1.
The Catma collapses code, amino-acid handedness and backbone handedness together — but whether that's a single simultaneous freeze or a late binary lock (L vs D) sitting on top of earlier foreclosures is not yet distinguished.
Closes with: independent dating evidence for each of the three components, showing whether they coincide or stagger. See Chain 2.4.
The persistence-through-discontinuity trait class turns out to sit inside one cell of a wider cost-sign partition — costly-to-hold versus costly-to-remove — from a companion paper on ferality. A costly-to-remove capacity would be retained through every discontinuity too, and would look identical in the record, without ever having been selected for it.
Closes with: a removability test independent of retention — genomic context, operonic structure, pleiotropy, knockout fitness in a permissive environment — run on capacities that are cheap to delete and retained anyway. See Chain 3.5.
Biotic molecules are roughly four times more robust to reaction-rule deletion than abiotic ones — but biotic molecules are also simpler (fewer functional groups, more thermodynamically stable), and simpler molecules have more synthesis routes almost by construction. The asymmetry might be measuring that instead.
Closes with: re-running the robustness comparison matched on functional-group count, molecular mass and network generation. Runs entirely on Wołos et al.'s already-public data — no new chemistry needed, and either outcome is publishable. See Chain 4.3.
Three soft spots are already named rather than hidden. Endogenous extinction needs its population-genetics defence actually stated, not assumed. "Cattening leaves a universal signature" is circular until it's broken by an independent diagnostic that could, in principle, fail to find one. Biology as the sole record of deep history needs to cross-validate against something outside biology somewhere, or the whole method is unfalsifiable by construction.
Closes with: stating each defence explicitly enough that it could fail — which is the whole point of writing them down here rather than leaving them implicit. See Chain 5.3 and /method/ §7.