The Methodology Reference

Biology Buildout — Wave 5 Findings + Light Anchor Identification + Stage 2 Synthesis Sketch

Wave 5 additions (5 Mns):

Corpus state: N=54 instance Mns (was 49). Validation: 310/312 (2 pre-existing topology failures unchanged).

Strict threshold (8/9) — 11 spines, 9 singletons, 45 of 54 covered (83%)

SpineNMembersNote
spine_08apis-mellifera, aplysia, c-elegans, drosophila, hypsibius, macrotermes, octopus, planariaBilaterian invertebrates RETURNED to N=8: tardigrade resolved c-elegans's singleton + the Wave 4 migrations reversed.
spine_18castor, chimpanzee, heterocephalus, human, monodelphis, mus-musculus, orca, vertebratesMammals (unchanged)
spine_27callorhinchus, corvus, gallus-gallus, lamprey, python-regius, xenopus, zebrafishNon-mammalian vertebrates (unchanged)
spine_35arabidopsis, dryopteris, oryza-sativa, pinus, selaginellaVascular plants (unchanged)
spine_45hydra, lumbricus, mnemiopsis, nematostella, strongylocentrotusBasal-eumetazoan + Lophotrochozoan-ND2 spine GREW to N=5. Spans 4 phyla — cnidarian + ctenophore + echinoderm + annelid. Convergent "intermediate-elaboration bridge" structural region.
spine_52amphimedon, trichoplaxMinimal-animal-grade pair (consistent)
spine_62anthoceros, physcomitriumBryophyte pair at strict 9/9 — hornwort + moss. Marchantia dropped out.
spine_72coprinopsis, schizophyllumBasidiomycete pair at strict 9/9 — resolves coprinopsis's persistent-singleton status.
spine_82ecoli, methanococcusProkaryotes
spine_92monosiga, salpingoeca-rosettaChoanoflagellates
spine_102paramecium, tetrahymenaCiliates

Singletons at strict (9): acropora, chlamydomonas, dictyostelium, halobacterium, marchantia, neurospora, rhizopus, volvox, yeast.

Moderate threshold (6/9) — 7 spines, just 2 persistent singletons (marchantia, yeast)

Key Wave 5 findings

Finding 1 — Migrations from Wave 4 reversed when tardigrade was added

c-elegans returned to the bilaterian-invertebrate spine; planaria returned; the bilaterian spine grew to its largest size (N=8). Adding a tardigrade (eutelous ecdysozoan with similar ND2 + bridge profile to c-elegans) rebalanced the spine centroid such that simpler-bridge bilaterians re-fit.

Methodology lesson confirmed: Wave-4 migrations were corpus-state-dependent artifacts, not permanent reassignments. The framework is reversible under corpus expansion — adding members that fit one end of a structural-region range pulls similar members back in.

Finding 2 — Three predicted pairs all confirmed at strict 9/9

Disciplined calibration (sample 3-5 neighbors before authoring) has now produced 13 of 13 hypothesis confirmations across Waves 2-5. The methodology rule developed after Wave 1 has held perfectly.

Finding 3 — Marchantia migrated to singleton when hornwort joined physcomitrium

Marchantia and physcomitrium were a moderate-threshold pair through Waves 2-4. With anthoceros added, anthoceros + physcomitrium paired at strict 9/9 — and marchantia dropped to strict singleton. Both at strict and moderate.

Diagnosis: marchantia's PF1 (no axial sporophyte) vs physcomitrium's PF3 + anthoceros's PF3 means marchantia is the structural outlier within bryophytes — the lower-elaboration liverwort gametophyte-dominant body plan. This is a real structural finding, not an authoring artifact: marchantia is the most-derived-reduced bryophyte at the partial-level grain.

Finding 4 — Rhizopus didn't pair with neurospora as predicted

Rhizopus + neurospora were both ND0 filamentous fungi with similar profiles. They didn't pair at strict. Both joined the moderate-threshold convergent zone with amphimedon + coprinopsis + schizophyllum + trichoplax.

The "fungal cluster" doesn't exist as a single attractor. Fungi split between:

This is a real structural finding: at the partial-level grain, fungi don't form a single cluster. The fruiting-body-vs-not distinction registers more strongly than the fungal-kingdom phylogenetic identity.

Finding 5 — The basal-eumetazoan spine grew across 4 phyla

spine_4 at strict threshold (hydra + lumbricus + mnemiopsis + nematostella + strongylocentrotus) is structurally heterogeneous: cnidarian (2x) + ctenophore + echinoderm + annelid. They share an "intermediate bridge profile" — neither at minimal-multicellular grade (amphimedon-trichoplax) nor at the elaborated-bilaterian grade (spine_0). This is a convergent middle-elaboration structural region spanning deeply-divergent phylogenetic groups.

Light anchor identification (no full JSON authoring)

Based on the Wave 5 strict-threshold spine inventory, proposed biology anchors:

Strong inductive anchors (5)

  1. mammals (N=8): castor, chimpanzee, heterocephalus, human, monodelphis, mus-musculus, orca, vertebrates. Substrate G5, Mem5. Includes eusocial, marsupial, placental, cetacean. Cohesion 9/9 strict.

  2. non-mammalian-vertebrates (N=7): callorhinchus, corvus, gallus-gallus, lamprey, python-regius, xenopus, zebrafish. Substrate G4/P5/Reg5. Spans agnathan → chondrichthyan → bony fish → amphibian → reptile → bird. Cohesion 9/9 strict.

  3. elaborated-bilaterian-invertebrates (N=8): apis-mellifera, aplysia, c-elegans, drosophila, hypsibius, macrotermes, octopus, planaria. Substrate G4/T4/R4/P4/Reg4/Mem4. Includes protostome bilaterians spanning nematode/arthropod/mollusk/platyhelminth/tardigrade. Cohesion 9/9 strict.

  4. vascular-plants (N=5): arabidopsis, dryopteris, oryza-sativa, pinus, selaginella. Substrate G4/T4/R4/P4/Reg4/Mem4 + bridge identical across members. Cohesion 9/9 strict.

  5. basal-eumetazoan-intermediate-bridge (N=5): hydra, lumbricus, mnemiopsis, nematostella, strongylocentrotus. Convergent structural region spanning 4 phyla (cnidarian + ctenophore + echinoderm + annelid). Cohesion 7/9 min. This is the highest-information anchor — it's a true convergent-structural finding.

Pair-anchors (6)

  1. minimal-multicellular-animal (N=2): amphimedon, trichoplax. Cnidarian + placozoan. The "basal animal" attractor.
  2. non-vascular-bryophyte (N=2): anthoceros, physcomitrium. Hornwort + moss. Marchantia excluded as outlier.
  3. fruiting-body-basidiomycete (N=2): coprinopsis, schizophyllum.
  4. choanoflagellate (N=2): monosiga, salpingoeca-rosetta.
  5. ciliate (N=2): paramecium, tetrahymena.
  6. prokaryote (N=2 strict / N=3 moderate): ecoli, methanococcus (+ halobacterium at moderate).

Moderate-threshold-only anchors (1)

  1. colonial-multicellular-small-genome-eukaryote (N=3 at moderate): chlamydomonas, dictyostelium, volvox. Cross-lineage convergent zone (green algae + Amoebozoa).

Persistent singletons (3)

Total: 12 anchor candidates — comparable to entity-arrangement Phase 1C (10 anchors + a few declaratives). Authoring them as JSON is mechanical given this inventory.

Stage 2 synthesis — early sketch

The Landscape Bootstrap Methodology doc has substantive content to draw on. The following rules are established (each with evidence-anchor):

Rules for landscape buildout

Rule 1 — Don't cluster at low corpus density

Anchor-authoring before ~30-40 Mns produces an inventory based on which categories happened to have 3 members in the early corpus, not on what the domain structures around. Evidence: biology spot-check at N=21 (Wave 0) surfaced 5 "anchor candidates" each with 3 members — but cluster shape at N=33 (Wave 2) substantially refined the picture, and the final inventory at N=54 (Wave 5) bears only partial resemblance to the Wave-0 candidates.

Rule 2 — Disciplined calibration before authoring

Sample 3-5 existing structurally-similar Mns + identify the tier the new Mn occupies + score it AT the tier unless there's a primitive-specific reason to differ. Evidence: Wave 1 (un-calibrated): 1 of 5 hypotheses confirmed. Wave 2-5 (calibrated): 13 of 13 hypotheses confirmed. The discipline turns the methodology from "buildout where authoring artifacts dominate" into "buildout where clustering is informative."

Rule 3 — Use the cluster output as a gap-finder, not a category lookup

Singletons in the output show where the corpus is sparse. Adding structural neighbors to those singletons specifically is the right next step. Evidence: Each wave's plan was driven by the previous wave's singleton inventory; this iterative gap-filling converged to a stable spine structure in 4 waves.

Rule 4 — Threshold sweeps reveal more than single-threshold reports

Reporting meta-stability at multiple thresholds (strict 8-9/9, moderate 5-6/9) reveals different structure. Strict captures core clusters; moderate captures convergent-structural zones across phylogenetically-distant lineages. Evidence: all five "convergent-structural zones" surfaced in biology (basal-eumetazoan-ND2, minimal-multicellular, colonial-eukaryote, etc.) appear at moderate but not strict thresholds.

Rule 5 — Singleton triage has four classes

(refined from gap-analysis-doc):

  1. Coverage artifact — resolves when a structural neighbor is added at moderate threshold
  2. Authoring artifact — calibration inconsistency on 1-2 primitives prevents intuitive pairing
  3. Structural attractor — genuinely isolated; persists even with structurally-similar neighbors added
  4. Outlier of category — the intuitive category is not actually a structural cluster (e.g., "protists" don't form one; "fungi" don't form one)

Evidence: all 4 classes documented in biology buildout. Examples: amphimedon was structural attractor (Wave 1-2) until trichoplax revealed it as part of a 2-member structural region (Wave 3-4); marchantia is now outlier-of-category (low-elaboration bryophyte); acropora is structural attractor (biomineralization-driven); the "fungi" category is outlier-of-category.

Rule 6 — Cluster membership is corpus-state-dependent

Existing Mns can MIGRATE under corpus expansion. Evidence: strongylocentrotus migrated from bilaterian-invertebrate (Wave 2) to basal-eumetazoan-ND2 (Wave 3) to planaria-pair (Wave 4) back to basal-eumetazoan (Wave 5). c-elegans dropped to singleton (Wave 4) and returned to spine (Wave 5).

Implication: anchor authoring should be done on a corpus that has been STABLE across at least 2 consecutive waves. Authoring at a mid-buildout state crystallizes a transient configuration.

Rule 7 — Pair-tightening also occurs under expansion

Sometimes more data narrows clusters: 3-member moderate-threshold zones can become 2-member strict pairs + 1 singleton when corpus expands. Evidence: amphimedon-coprinopsis-trichoplax (Wave 3 moderate triple) → amphimedon-trichoplax strict pair + coprinopsis singleton (Wave 4) → coprinopsis-schizophyllum strict pair (Wave 5).

This is the reverse of "more data = better resolution" — sometimes more data exposes within-category variance.

Rule 8 — Substrate-grade dominates over behavioral/cooperative organization

The partial-level grain treats structural/developmental/cellular organization (which clusters) distinctly from behavioral/ecological/cooperative organization (which doesn't). Evidence: eusocial Mns (apis + macrotermes + heterocephalus) joined their substrate-determined spines, not a cross-phylum eusocial spine. Cultural-transmission Mns (orca, corvus) joined substrate spines. Extreme cognition (octopus, corvus) didn't pull out of substrate-determined spines.

Methodological implication: the framework is a structural decomposition, not a behavioral one. Anchors should be named for what they STRUCTURALLY share, not for what they behaviorally do.

Rule 9 — Cross-domain replication is the methodology validation

The same recipe (cluster_classical → stability → meta-stability across thresholds) produces coherent spine inventories in BOTH entity and biology arrangements. Evidence: silhouettes 0.30-0.55 in entity, 0.42-0.80 in biology (biology higher — more structured domain). Spine inventory shape comparable: 4-5 large spines + many small pairs.

The methodology generalizes; what differs across domains is which spines emerge as anchors, not whether the framework produces clusterable structure.

Rule 10 — Convergent-structural-zones at moderate threshold are a finding-type

When 3+ phylogenetically-distant lineages cluster at moderate threshold (not strict), that zone is a cross-lineage structural convergence finding — exactly what the partial-level grain is designed to surface. Evidence:

These zones are the convergent-evolution findings that justify the framework's structural claims.

What Stage 2 doc needs to do

A. Codify these 10 rules. Each rule with one paragraph of statement + 1-2 paragraphs of evidence + a "how to apply" line.

B. Document the buildout protocol — the wave-by-wave gap-analysis → authoring → re-clustering → findings cycle.

C. Document the singleton triage taxonomy with the 4 classes + diagnostic tests.

D. Capture cross-domain comparison — entity vs biology methodology fit.

E. Connect to existing canonical-architecture-strategy.md — Rules 1-10 here extend Rules 1-10 of the canonical strategy (which was about within-arrangement discipline). These are about cross-arrangement buildout discipline.

Estimated length: 800-1200 lines as a sibling doc, then merged into canonical-architecture-strategy.md as a new §11 "Landscape Bootstrap Methodology."

Files

PurposePath
This combined Wave 5 + anchor ID + Stage 2 sketchmethodology_strategy/biology-wave5-findings-and-anchor-id.md
Wave 5 Mn JSONsdata/manifestations/{nematostella,schizophyllum,rhizopus,anthoceros,hypsibius}.v1.json
Meta-stability latestoutput/results/cluster-meta-stability-biology.v1.json
Meta-stability figureoutput/figures/cluster-meta-stability-biology.png

Referenced by the model

Cited as a source by 15 model records (browse the model census):