NA-10 — The scale ladder: quark to galaxy, and the gradients between

What you are reading. The wing's spine. A rung-by-rung placement table across ~36 orders of magnitude in length, and — the actual payload — the rule for knowing when a number you fitted on one rung stops meaning anything on the next. Every value below carries a source, or is written NOT-MEASURED (no measurand exists) or NOT-SOURCED in this pass (a measurand exists; this pass did not fetch the primary). There is no fourth state — a plausible number with silent provenance is exactly the defect this chapter exists to catch — and where a re-read found one standing in the ladder table, it is named in the NOT-SOURCED rows below rather than left silent. Nothing here raises any UNI rung; a nature citation is never a UNI gate.

The span, and what it is not

From the proton's rms charge radius, 8.4075(64) × 10⁻¹⁶ m (CODATA 2022, NIST), to the Milky Way's D25 isophotal diameter, 26.8 ± 1.1 kpc ≈ 8.3 × 10²⁰ m (attributed to Goodwin et al. 1997/98, The Observatory 118:201–208, via secondary summary — see the open row), is a factor of ~10³⁶.

The ladder is not a continuum with things sprinkled along it. It is a set of regimes separated by boundaries, and the whole engineering value of the chapter is in the boundaries. Inside a regime, one force dominates, one or two dimensionless groups govern, and fitted power laws behave. Across a boundary, the dominant force changes identity and every fitted exponent you carried across is a category error wearing a decimal point.

The ladder

L = characteristic length. t = a characteristic time of that rung's own dynamics (see the honesty note on the time gradient below — these are not one measurand). Blanket = is a Markov blanket identifiable (cross-ref NA-04, mind/body/world)?

Read the "Blanket?" column downward. It is not monotone. Blankets appear at rungs 6, 7, 10 and vanish above and below. A galaxy is enormous and has no blanket. A mitochondrion is a micron across and has one. Size is not what makes a boundary.

What a rung actually is — emergence, stated operably

A new rung exists when all three hold. State them or you have not identified a level, you have merely named a size.

(a) A boundary with real conditional independence forms. Internal states are screened from external states by the boundary — this is the Markov-blanket condition (cross-ref NA-04). It is a factorization claim about the dynamics, and it is checkable. Not "there is a surface"; "the inside is conditionally independent of the outside given the boundary".

(b) A new characteristic timescale separates from the level below. This is the actual mechanism of hierarchy, and it is Simon (1962), "The Architecture of Complexity", Proc. Am. Philos. Soc. 106(6):467–482: near-decomposability means intra-component linkages are stronger than inter-component linkages, which has the effect of separating the high-frequency dynamics of the hierarchy — the internal structure of the components — from the low-frequency dynamics — the interaction among components. Timescale separation is not a description of hierarchy; it is what makes one. Simon's watchmaker parable supplies the evolutionary half: the watchmaker who builds stable ten-part subassemblies loses only the last subassembly when interrupted, and so his expected work loss is smaller by orders of magnitude. Hierarchies are what evolution can reach, because stable intermediates are what interruption spares.

(c) The level has its own dynamics that do not require tracking the level below. This is Anderson (1972), "More Is Different", Science 177(4047):393–396, DOI 10.1126/science.177.4047.393, rejecting the "constructionist hypothesis": the ability to reduce everything to simple fundamental laws does not imply the ability to start from those laws and reconstruct the universe. Anderson's claim is not mystical and does not need to be. It is that broken symmetry at each level generates concepts (rigidity, a phase, a species) that are not present in the level below.

Note the honest ordering: (b) causes (c). If the timescales did not separate, you would have to track the fast level and there would be no autonomous slow dynamics to write down. Emergence here is a statement about spectral gaps in a dynamical system, not about magic.

The best regime-change example there is: Reynolds number and the death of the scallop theorem

Re = ρvL/µ = vL/ν — the ratio of inertial to viscous forces. ν_water ≈ 1.0 × 10⁻⁶ m²/s at 20 °C.

Vogel's table of self-propelled organisms (Life in Moving Fluids: The Physical Biology of Flow, Princeton UP) runs from a bacterium at Re ≈ 10⁻⁵ (0.01 mm/s) to a large whale at Re ≈ 3 × 10⁸ (10 m/s), with a sea-urchin sperm at ~0.03 and a tuna at ~3 × 10⁷ in between. Check the endpoints independently: whale, v = 10 m/s, L = 30 m → Re = 3 × 10⁸ ✓. E. coli, v ≈ 3 × 10⁻⁵ m/s, L = 2 × 10⁻⁶ m → Re ≈ 6 × 10⁻⁵ ✓ (NA-08).

That is a span of ~10¹³, not ~10¹⁰. The ~10¹⁰ figure in circulation is short by about three orders of magnitude; the sourced span is ~13 orders. Corrected here rather than carried.

And the physics of swimming changes identity across it, not degree.

At Re ≈ 10⁻⁵, inertia does not exist. Compute the coasting distance for a 1 µm sphere when propulsion stops: Stokes drag gives a decay time τ = 2ρa²/(9µ) = 2(10³)(10⁻¹²)/(9 × 10⁻³) = 2.2 × 10⁻⁷ s, so the coast is d = vτ = 3 × 10⁻⁵ × 2.2 × 10⁻⁷ = 6.7 × 10⁻¹² m = 0.067 Å — about 1/16 of a hydrogen atom's diameter (2a₀ = 1.06 Å). Order-of-magnitude consistent with Purcell's own back-of-envelope in Purcell (1977), Am. J. Phys. 45(1):3–11, DOI 10.1119/1.10903. A bacterium cannot coast. There is no "let go and glide". Stop pushing and you have already stopped.

Hence Purcell's scallop theorem: a low-Re swimmer executing a reciprocal stroke — a shape sequence identical when reversed — nets zero displacement in an incompressible Newtonian fluid, because Stokes flow is time-reversible. Purcell's phrasing: it "exactly retraces its trajectory, and it's back where it started" (Purcell 1977). A real scallop, at Re ~10⁴–10⁵, swims perfectly well by clapping. Shrink it to a micron and the identical stroke transports it nowhere, forever. Nothing about the animal changed. The dominant term in the Navier–Stokes equation did.

And the theorem is premise-bound, which is the second half of the lesson. Qiu et al. (2014), "Swimming by reciprocal motion at low Reynolds number", Nature Communications 5:5119, DOI 10.1038/ncomms6119, built a single-hinge "micro-scallop" that does propel by reciprocal motion at Re = 1.4 × 10⁻⁴ – 3 × 10⁻³ — in shear-thickening and shear-thinning non-Newtonian fluids, with a Newtonian glycerol control. The theorem was not refuted; its premise ("Newtonian") was withdrawn. A law fitted inside a regime is a law about that regime's premises. Change a premise and you are on a different rung of a different ladder.

The other regime change: quarter-power holds inside a regime, and breaks at the boundary

Kleiber's B ∝ M^(3/4) is the wing's most-quoted ratio, and it is the wing's best worked example of a regime-bounded law.

Boundary evidence. DeLong, Okie, Moses, Sibly & Brown (2010), "Shifts in metabolic scaling, production, and efficiency across major evolutionary transitions of life", PNAS 107(29):12941–12945, DOI 10.1073/pnas.1007783107, fitted metabolic rate against body mass across ~16 orders of magnitude in size, split at the taxonomic boundaries:

Their conclusion, in their own framing: Kleiber's 3/4-power law does not apply universally across organisms. The exponent does not drift — it changes sign of curvature relative to linear, from superlinear to linear to sublinear, and it does so exactly at the major evolutionary transitions. That is not noise. That is a rung boundary showing up in the residuals.

Curvature evidence, inside the metazoan regime. Kolokotrones, Van Savage, Deeds & Fontana (2010), "Curvature in metabolic scaling", Nature 464:753–756, DOI 10.1038/nature08920: the mass–metabolic-rate relation has convex curvature on a logarithmic scale and is therefore not a pure power law at all, even after accounting for body temperature; it needs a quadratic term in log-mass, which breaks the scale invariance a power law asserts. The payoff is diagnostic: attempts to fit a straight line to what is really a curve produce "scaling exponents" that depend on which data you used — data sets dominated by small mammals tend to yield ~2/3, data sets dominated by large mammals tend to yield ~3/4. The decades-long 2/3-vs-3/4 argument was, on this account, two groups fitting straight lines to different arcs of one curve and each reporting their arc's slope as a universal constant.

*Fence: Kolokotrones et al. is contested, and the exchange has two published turns — carry both. MacKay (2011), "Mass scale and curvature in metabolic scaling", J. Theor. Biol. 280(1):194–196, is a published comment against it; Deeds, Savage & Fontana (2011), "Curvature in metabolic scaling: a reply to MacKay", J. Theor. Biol. 280(1):197–198, DOI 10.1016/j.jtbi.2011.03.036, is a published reply by three of the four original authors, in the same issue immediately following the comment. The curvature claim is carried as OBSERVED-CONTESTED — live on both sides, settled by neither. Citing the comment without the reply would be a one-sided rendering of a dispute this chapter's own method requires it to carry whole.*

The regime-change rule (the operable output)

A power law fitted within a regime does not extrapolate through a regime boundary. Not "less accurately". It does not carry, and its exponent stops being a fact about the world.

Detection procedure. Run all five before extrapolating a fitted exponent across any decade you did not measure:

1. Did the dominant dimensionless group move by orders of magnitude across the span? (Re: ~13 orders, bacterium → whale.) If yes, suspect a boundary.
2. Did the dominant force change identity? (Viscosity → inertia. Diffusion → convection. EM → gravity.) A change of dominant term, not of coefficient, is a boundary.
3. Does the log-log plot have curvature? Fit a quadratic in log-mass and test the quadratic coefficient against zero (Kolokotrones et al. 2010). Curvature is a power law's own confession.
4. Does the fitted exponent depend on which subset of the data you fit? If small-dominated and large-dominated subsets give different slopes, you have a curve, not a law (Kolokotrones et al. 2010: 2/3 vs 3/4).
5. Is there an independent structural boundary at the suspected break? (DeLong et al. 2010: prokaryote → protist → metazoan.) An independently-motivated break is worth far more than a break found by searching for one — a break located only by scanning residuals is a fitted artefact until pre-registered. Per M2, the verdict is the CI bound excluding the threshold, never the point estimate.

The gradients: which are measured, which are rhetorical

This section is the chapter's honesty fence. The ladder invites gradient talk, and most gradient talk is unearned.

MEASURED, real, comparable across rungs:

• Length. ~10³⁶ — and it is the one column whose endpoints are commensurable to within a factor of ~2, which is nothing across 36 orders. It is still not literally one measurand, and this chapter does not get to make the claim it refuses t: rungs 1/2/13/14 are radii, rung 16 is a photometric isophotal diameter (D25 — the isophote where B-band surface brightness falls to 25 mag/arcsec², a brightness convention rather than a geometric extent), and rungs 4–12 are ranges. The ladder's top rung is therefore measured by a different kind of instrument than its bottom rung. The span is real; the uniformity is approximate. Unlike t, the heterogeneity does not change the physical question being asked — a factor of ~2 and a radius-vs-diameter convention do not move a 10³⁶ conclusion — which is why L stays MEASURED and t does not.
• Specific power (W/kg). The same measurand at every rung, and it is worth stating because it runs the opposite way to intuition. Both endpoints are conversions, and both are labelled so. An E. coli runs at ~10³ W/kg — computed, not measured: BNID 109687 is an O₂ uptake rate (30 mmol O₂/gDW/h, E. coli strain C-3000, glucose minimal medium), turned into W/kg through the enthalpy of O₂ consumption and a cell mass. It is not a W/kg observation. The Sun runs at L⊙/M⊙ = 3.828 × 10²⁶ W / 1.988 × 10³⁰ kg = 1.9 × 10⁻⁴ W/kg (computed from IAU 2015 nominal values). A bacterium outputs roughly 5 million times more power per kilogram than the Sun does. The ratio is MODELED ÷ MODELED — the same arithmetic on both sides, so the same class on both sides, which is the point: dividing a sourced luminosity by a sourced mass and dividing a sourced O₂ rate by a sourced mass cannot be two different classes in one comparison. The conclusion survives the one open scope question in the chain (dry- vs wet-mass basis, worth ~4× — see the row), because the result is a factor of ~10⁶ and the ambiguity is a factor of ~4. It demolishes any "energy density rises with scale" narrative with room to spare.
• Reynolds number — but only within the self-propelled-body domain, rungs 7–10. It is not defined for an atom or a galaxy.

PART-MEASURED, PART-RHETORICAL:

• Time. Each rung's t is measured. But r_p/c, a C–H stretch period, a cell division time, and the Sun's nuclear lifetime are not one measurand — they are a light-crossing time, a vibrational period, a doubling time, and a fuel-exhaustion time. The endpoints are real; the gradient is an artefact of putting four different physical questions in one column. Column t above is a placement aid, not a scaling relation, and must never be regressed.

RHETORICAL — NOT-MEASURED, and the fence is here:

• "Information density." No agreed measurand exists spanning quark to galaxy. Genome information content is measurable at one rung; there is no bits-per-m³ figure comparable between a nucleon and a galaxy. NOT-MEASURED. Falsifier: state the measurand, state the units, measure two rungs with it.
• "Degree of internal model." No measurand. No units. No instrument. This is the gradient the ladder most invites and it is the one with the least behind it. Recorded NOT-MEASURED, not softened. Whatever a variational scheme is doing at rung 7, no one has published a scalar that orders rungs 1–16 by "modelledness". Writing the ladder as an ascent toward mind is a rhetorical move dressed as a measurement, and it is exactly the kind of move this wing exists to catch.
• "Complexity." Same verdict. Simon (1962) gives a structural criterion (near-decomposability) that is checkable per system; it does not yield a scalar that monotonically increases with size.

Major transitions as rung-crossings

Maynard Smith & Szathmáry (1995), The Major Transitions in Evolution, Oxford University Press, ISBN 978-0-19-850294-4 (and Szathmáry & Maynard Smith 1995, Nature 374(6519):227–232, DOI 10.1038/374227a0) list eight:

1. Replicating molecules → populations of molecules in compartments
2. Independent replicators (RNA) → chromosomes
3. RNA as gene and enzyme → DNA genes; protein enzymes
4. Prokaryotes → eukaryotes
5. Asexual clones → sexual populations
6. Protists → multicellular organisms
7. Solitary individuals → colonies with non-reproductive castes
8. Primate societies → human societies with language

Several of these are rung-crossings in exactly the sense above: entities that previously replicated independently became bounded inside a new blanket with its own slower dynamics. Transition 1 builds the rung-7 blanket. Transition 4 builds rung 6 (the mitochondrion's blanket is a former rung-7 blanket demoted to an organelle's). Transition 6 builds rungs 8–10. And DeLong et al. (2010) is the receipt that these are physically real boundaries and not just a taxonomist's convenience: the metabolic exponent changes at them.

The framework is influential and it is debated — carry both. West, Fisher, Gardner & Kiers (2015), "Major evolutionary transitions in individuality", PNAS 112(33):10112–10119, DOI 10.1073/pnas.1421402112, break transitions into two steps (group formation; transformation into an integrated entity) and argue explicitly toward "a simpler and more unified description" — the framing itself implying the original list offered a mixture of explanations rather than one. A recurring criticism holds that the list lacks theoretical unity, and that transition 8 (language) does not meet the criterion of previously free-living entities becoming integrated into a higher-level individual. This criticism is recorded here from secondary summary; the primary sources for the theoretical-unity objection were NOT located in this pass — see the open row. Do not cite the objection to a specific author on this chapter's authority.

Place any object on the ladder — the 5-step procedure

1. Measure L. Take the characteristic length — the shortest dimension that carries the relevant flux, not the longest dimension you can find. (NA-08: Thiomargarita is 750 µm wide and its L is 1–2 µm, because the cytoplasm is a shell. Getting L wrong puts the object on the wrong rung and every later step inherits the error.)
2. Name the dominant force. Strong / EM / viscous / inertial / gravitational. If two contend, you are at a boundary — say so and stop extrapolating, do not average.
3. Compute the governing dimensionless group(s) from the table's column for that rung. Re and Pe at rungs 7–10; α at rung 2; α_s at rung 1; Ra/Ro at rung 13. The group, not the size, tells you which physics is running.
4. Test for a blanket. Can you state what is inside, what is the boundary, and what is outside — with conditional independence, not just a visible surface? Yes → it is an entity at that rung. No → it is a region, not a system (NA-08's rule; NA-04's typing). Rungs 12–16 fail this test and that is fine; galaxies do not need blankets.
5. Check the regime before importing any ratio. Run the five-test procedure above against every fitted exponent you were about to carry in. If any of the five fires, the ratio does not cross — record it NOT-APPLICABLE at the new rung rather than extrapolating. Then, per M7: any design taken from this placement must still beat a tuned baseline on a pre-registered metric with a discriminator that collapses the gain, or it is recorded NEGATIVE.

The numbers

Falsifier (operable)

The chapter's central structural claim is the regime-change rule: a power law fitted within a regime does not extrapolate through a regime boundary, where a boundary is marked by a change in the dominant force and an orders-of-magnitude change in the dominant dimensionless group.

It is refuted by exhibiting one power law, fitted on data entirely within one regime, that predicts — within its stated CI, on data never used in the fit — observations on the far side of a boundary at which the dominant force demonstrably changed identity. Concretely: fit B ∝ M^b on metazoans alone and predict prokaryote metabolic rates within CI. DeLong et al. (2010) is the standing evidence that this fails (0.76 vs 1.7). A single clean success moves the chapter.

Secondary, row-local falsifiers are in each table row. A refuted row moves that row. A refuted regime-change rule moves the chapter.

Recorded INADMISSIBLE / NEGATIVE (first-class, inline)

• "Re changes by ~10¹⁰ from bacterium to whale." — NEGATIVE, corrected here. The sourced span (Vogel's table, endpoints independently recomputed) is ~10¹³, ~13 orders, not ~10. Recorded because the wrong figure was carried into this chapter's own brief and would have propagated. The correction does not weaken the point; it strengthens it.
• "The scale ladder is a ladder of increasing consciousness / awareness / mind." — INADMISSIBLE. No measurand, no units, no instrument, no falsifier. The blanket column above is not monotone in size (rungs 6, 7, 10 have blankets; 12–16 do not), so even the one structurally checkable property does not ascend with scale. Recorded, not mocked: the intuition that the ladder points somewhere is honest and common. It is simply not a measurement, and dressing it as one is the defect.
• "The Earth/biosphere is an organism" (Gaia, as a literal claim). — INADMISSIBLE as stated at rung 13. It fails criterion (a): no boundary with demonstrated conditional independence is exhibited. Gaia as a heuristic about coupled biotic–abiotic feedback names real, measurable couplings and is admissible in that form. The organism claim, as stated, names no falsifier. The two must not travel together.
• "Emergence means the whole is more than the sum of its parts." — INADMISSIBLE as stated. Unfalsifiable: no observation is specified that could refute it. The admissible neighbour is in the body of this chapter and has three checkable conditions (conditional independence; timescale separation; autonomous slow dynamics), each with a test. Anderson (1972) is the earned version; the slogan is the unearned one. The distance between them is the method.
• NEGATIVE / extrapolation trap: quoting Kleiber's 3/4 as a law of life. It is a metazoan-regime fit. Prokaryotes are superlinear at ~1.7 (DeLong et al. 2010). Recorded because the 3/4 exponent is the single most over-extrapolated number in this wing.
• NEGATIVE / conflation trap: treating the t column of the ladder table as a scaling relation. It is a placement aid built from four different physical measurands. Regressing it produces a number about nothing.
• NEGATIVE / averaging trap — and the method arrived. In 2016 R₀ (Sun → Galactic Centre) stood at 8.32 ± 0.07(stat) ± 0.14(sys) kpc (Gillessen et al. 2017, ApJ 837:30) or 7.86 ± 0.14(stat) ± 0.04(sys) kpc (Boehle et al. 2016, ApJ 830:17) — ~2.2σ apart on errors combined in quadrature (computed here), intervals not overlapping. Averaging them would have manufactured a value no one measured. The row's instruction was resolve by method, and the method then resolved it: GRAVITY Collab. 2019 (A&A 625:L10) measured R₀ geometrically — a different instrument, not a re-weighting of the old two — at 8178 ± 13 ± 22 pc (0.3%), landing near Gillessen, and located where the disagreement came from: Boehle's S2-only fit gives 8.02 ± 0.36 ± 0.04 kpc, "completely consistent" with GRAVITY, so the offset lived in the combined S2+S38 fit rather than in the data. Kept as this chapter's worked example precisely because it resolved, and it resolved in the shape the rule predicted: carrying both values intact is what left the discrepancy legible as a methodological signal. An average would have erased the very thing that got fixed. The rule made a prediction about how the standoff would end, and the prediction held — which is why a resolved case is a stronger teacher here than a live one.
• NOT-SOURCED in this pass: Vogel's Re table read from the primary; the IAU 2012 AU definition primary; the Goodwin et al. Milky Way D25 primary; trophic transfer efficiency; O₂ tissue penetration depth; aortic Womersley number; the primary source for the MTE theoretical-unity objection. Each is flagged in its row with a closure action. None of them is load-bearing for the regime-change rule.
• NOT-SOURCED in this pass — ladder-table values carrying no row in "The numbers". Caught by this chapter's own gate on re-read, and named here rather than left standing, because the header's promise was not true of the chapter's own table: C–C bond length ~1.5 × 10⁻¹⁰ m (rung 3 — the bond energy has a row, the length did not); aortic Re ~10³–4×10³ (rung 9 — the Womersley α in the same cell was flagged NOT-SOURCED, the Re sitting beside it was not); galactic year ~7 × 10¹⁵ s / ~225–250 Myr (rung 16); heliopause ~120 AU (rung 15); mitochondrion L 5 × 10⁻⁷–10⁻⁶ m (rung 6); E. coli division time ~1.2 × 10³ s (rung 7). None is fabricated — all are standard textbook values, and none is load-bearing for the regime-change rule — but standard-and-probably-true is not the same as sourced, and a plausible number with silent provenance is the precise defect this chapter exists to catch. Recorded because the gate caught its own author, which is the only evidence that a gate works. Closure action for each: fetch a primary (a cross-ref to NA-08 is an acceptable source; silence is not) and give it a row with a falsifier.
• NOT-MEASURED: any scalar ordering rungs 1–16 by "information density", "degree of internal model", or "complexity".

HONEST FENCE — MODELED

Fenced MODELED. Individual rows carry their own classes (many OBSERVED-REPLICATED, several OBSERVED-CONTESTED, several NOT-MEASURED/NOT-SOURCED), but the chapter as an artifact is a placement model: it composes measured constants and published fits through stated assumptions — that "characteristic length" is well-defined per rung; that one force dominates per regime; that adopted IAU nominal constants stand in for measured stellar/planetary values (they are exact by adoption, explicitly not current best estimates, and must never be quoted as measurements with uncertainties). The assumptions are the fence. The rung count and the rung boundaries are a choice, not a discovery: a different carving (adding virus, biofilm, galaxy cluster) is equally defensible. What is not a choice is that the dominant force changes identity across the span, and that fitted exponents do not survive those changes.

Per Gould & Lewontin (1979), "The Spandrels of San Marco and the Panglossian Paradigm": nothing above establishes that any rung's arrangement is an optimum. Drift, phylogenetic inertia, developmental constraint, and frozen accidents produce structure that solves nothing. Nature's authority here is precisely and only this: it already ran the experiment under real constraints, with the failures deleted — so convergence at a rung is evidence of a constraint-optimum, and every number above is a hypothesis generator, never a proof. Per M7, a design taken from this ladder must beat a tuned baseline on a pre-registered metric with a discriminator that collapses the gain and a true computed-residual ablation, or it is recorded NEGATIVE. Per M22, this chapter's own placements are an upstream prior and must never be re-served as fresh evidence downstream.

Not claimed

• Not claimed: that the ladder is a progression, a hierarchy of value, an arrow, or a direction. It is an ordering by length. The blanket column does not ascend with it, and specific power runs against it by ~5 × 10⁶ from bacterium to Sun.
• Not claimed: that "degree of internal model" or "information density" increases along the ladder. NOT-MEASURED. No measurand exists. This is the gradient the ladder most invites and the one with the least behind it, and that asymmetry is stated on purpose.
• Not claimed: that emergence is mysterious, or that it is trivial. The operable version has three testable conditions and is a claim about spectral gaps in a dynamical system. Both the mystical reading and the dismissive reading are defects.
• Not claimed: that the Maynard Smith & Szathmáry framework is settled. It is influential and debated; the debate is carried in the body and the objection's primary source is NOT-SOURCED in this pass.
• Not claimed: that Kolokotrones et al. (2010) settles metabolic scaling — nor that MacKay settles it against them. It is OBSERVED-CONTESTED and the exchange is live on both sides: MacKay (2011), J Theor Biol 280(1):194–196, is a published comment against it; Deeds, Savage & Fontana (2011), J Theor Biol 280(1):197–198, is the original authors' published reply in the same issue. Both turns are carried; neither is the verdict. The regime-change rule does not rest on it — DeLong et al. (2010) carries the boundary claim independently, and the rule would stand if the curvature claim fell.
• Not claimed: that the sixteen rungs are the right sixteen. The carving is a choice; the boundaries' physics is not.
• Not claimed: that any citation above raises any UNI rung. A nature citation is NEVER a UNI gate. The NATURA twelve-value class (OBSERVED-REPLICATED / OBSERVED-SINGLE / OBSERVED-CONTESTED / MODELED / MODELED-CONTESTED / HYPOTHESIZED / INADMISSIBLE / SUPERSEDED / NOT-MEASURED / NOT-SOURCED / NOT-CONFIRMED / NOT-LOCATED — six of them registered by NA-00 amendment 2026-07-15-A) and the UNI four-value fence (proven / designed / hypothesized / not-yet-built) describe different kinds of claim and never merge. This chapter contains zero UNI claims.
• QUAESTIO-APERTA: "full human" and "beyond human" appear nowhere here as a target, milestone, or deliverable. They are permanent open questions. The scale ladder has no top, no destination, and no rung labelled with either of them — and the temptation to read one in at rung 10 or 11 is exactly what the NOT-MEASURED fence on "degree of internal model" exists to refuse.