What is the thing that survived?

Biology treats "all life shares one ancestor" as a settled preface, then moves on to what's interesting. The bioblob is what happens if you take the preface literally instead — not a metaphor for common descent, but a claim about what kind of object is actually sitting there once you do.

The Hadean wasn't a difficult environment; it was a succession of environments, each lethal to something optimised for the last.◦ So FUCA wasn't selected for being good at living in the Hadean — it was selected for still existing after the Hadean stopped being whatever it currently was. A poly-extremophile is not adapted to an extreme. It is adapted to the transition between extremes.

Take a tree and cut it through the crown. What the cut face shows is a scatter of separate circles, distributed across a plane with empty space between them — and nothing in that picture says tree. To see that they're one object you have to step out of the cutting plane; once you do, the empty space between the circles turns out to be where the connection was. The cut doesn't merely hide the fourth dimension. It manufactures plurality. Biology is the study of cross-sections — ecology, physiology, most of genetics performed on a slice a few decades thick through an object four billion years long, a ratio of about one part in 10⁸.

The objection is exactly that kind of slice error. A phylogenetic tree is not a diagram of relations between separate objects; it is a map of the internal structure of one object, drawn as though from outside it. And the Gaia comparison gets sharper, not softer: Gaia needs selection on the whole, which it cannot have. This needs selection in the whole, which every tree and every body already has. Extinctions are leaf-fall, not deaths — which is why "100% successful for four billion years" is a structural observation rather than survivorship bias.

The spore concession. A tun does zero dissipation and is unambiguously not dead, so it must be carrier-only — record held, generator and filter both stopped. That concedes that what life is being defined as isn't running in something we'd still call alive, worth conceding deliberately rather than discovering under review.

The matrilineal consequence. If branching requires cytoplasm and sperm supply none, the manifold branches only through eggs, and organelle inheritance stops needing a separate explanation.

The obvious prediction — that generalists show lower peak performance than specialists — is the jack-of-all-trades hypothesis, and it's wrong: a phylogenetically informed comparison of 28 teleost species found peak aerobic capacity unlinked to thermal performance breadth, and a niche-breadth review concludes performance-breadth trade-offs may not be common determinants of niche breadth at all.◦ Withdrawn That original bioblob prediction is withdrawn — full reasoning in /claims/.

The corrected prediction is about loss of apparatus, not degradation of function: across independent transitions from free-living to obligate specialist, genes encoding persistence capacities should be lost preferentially — earlier than genes of comparable expression cost serving within-regime function. The obvious confound is serious: genome reduction in endosymbionts is usually attributed to drift, not selection. Distinguishing them is the experiment — preferential order and scaling with environmental stability, rather than with effective population size, are separable predictions.

Palaeontology already shows the cost profile directly: in Jablonski's Late Cretaceous mollusc data, background-time survival predictors (larval development, geographic range, species richness) go dark during the extinction event itself, and what predicts survival instead is clade-level geographic deployment — 55% of widespread bivalve genera survived against 9% of endemics. Anstey's Palaeozoic Bryozoa show the same pattern from the opposite side: with mass extinctions removed, morphologically complex (specialist) taxa had lower background extinction than simple generalists. Specialists are better in stable times and die at transitions — the class's defining property, independently, in a different phylum.