. . . . "Downloaded Meshes.zip from Zenodo 2593861 and the measurement tables. Specimen identifiers are matched case-insensitively, and macOS resource-fork members whose names start with '._' are skipped. For each internal chamber mesh (SampleID-CH#.stl, numbered from the final chamber) and the shrink-wrapped whole test (SampleID-WRAP.stl), up to 5000 vertices are sampled and an algebraic sphere is fit. The specimen score is the median normalized RMS residual of the chambers, excluding the earliest, minus the whole-test residual. A one-sided Wilcoxon test across specimens is compared with the pre-registered threshold. The isoperimetric quotient from the measurement table is computed and not used, because the meshes are present." . . "Measured: -0.008931936051936817 median across specimens of (median chamber sphere-fit residual minus whole-test residual); negative means chambers are closer to a sphere (n = 37, one-sided Wilcoxon p (chambers closer to a sphere) = 0.03262675827863859). Interpretation: Thompson's 'figure of equilibrium shaped by surface tension' for a new chamber is read as a spherical segment, the closed surface of constant mean curvature available from a mesh. The comparison is to the whole test, which the same sentence says is not itself a figure of minimal area. Globigerina and Rotalia are the named examples; planktonic foraminifera in the archive are in that class of multichambered shells, and any absence of Rotalia or of desmids is a limit on scope, not a reason to drop the specimens that are present. The moment of solidification is not in a mesh of the finished shell." . "Claim test of thompson1917-ch05-08: supports, confidence 0.35" . . . "Segmented per-chamber surfaces, or per-chamber volume and surface area, for multichambered foraminifera, with a whole-test surface or measurement for the same specimen and a way to tell chambers apart. Public and licensed for reuse. Desmid semicells and live images of chamber formation would be admissible for the clauses this test leaves out; they are not required for the geometric reading." . "claim-test/0.1.0" . "biological" . "Chamber surfaces depart from a sphere by as much as, or by more than, the whole multichambered test." . "A sphere fit on the finished chamber mesh, including the septal wall, is a geometric comparison. It does not show that surface tension shaped the chamber." . "Globigerina bulloides, the named genus present in the archive, has median paired difference -0.06252824301103344 (9 of 10 negative). Other genera in the same table are near zero or positive. The verdict uses all 37 specimens, not Globigerina alone." . "No Rotalia and no desmids are in this archive. The shells are finished calcite. Nothing here records a chamber while it was fluid or the moment it solidified, so the surface-tension mechanism and the fluid-to-rigid timing are untested." . "Six specimen ids in the chamber table (RS11-5, RS11-6, RS12-1, RS12-3, RS14-3, T1) have no joined shrink-wrap whole-test mesh and are absent from the test. One mesh id has no species label in the chamber table." . "The isoperimetric quotients from the volume and area table are not the verdict. Some chamber quotients exceed 1, which a closed surface cannot, so those columns are not a clean sphere comparison. Their paired test does not meet the support rule." . "The sphere-fit difference is small. Median normalized RMS residual is 0.12088461729839131 for chambers and 0.13400139724767005 for the whole test. The median paired difference is -0.008931936051936817 (mean -0.014913645970034601). 23 of 37 specimens have chambers closer to a sphere than the whole test (one-sided Wilcoxon p 0.03262675827863859)." . "Individual chamber surfaces are closer to a fitted sphere than the surface of the whole test from the same specimen." . "Pre-registered. REFUTES if n_specimens >= 10 and the Wilcoxon test is significant (p < 0.05) in the opposite direction (chambers farther from a sphere than the whole shell). MIXED if one of the majority-sign and the significance requirements holds and the other does not. Inconclusive, reason underpowered, if fewer than 10 specimens can be measured; data-unavailable if neither meshes nor area-and-volume tables can be fetched; not-operationalizable if chamber order or a closed surface cannot be identified. Desmids and the fluid-to-rigid timing are out of scope and are not scored. A sphere fit on a rigid shell does not show that the chamber was a fluid equilibrium when it formed. Thresholds are not revised after the meshes are opened." . "0.35"^^ . . "supports" . "Each foraminiferal chamber, once formed, lies closer to a sphere than the whole multichambered shell does." . "Paired one-sided Wilcoxon test across specimens of the difference between the median chamber sphere-fit residual and the whole-test residual, p < 0.05, n >= 10, with the majority of specimens in the predicted direction. The isoperimetric quotient is the fallback when meshes are not in the archive." . "Pre-registered geometric reading. For each specimen that has a whole-test mesh and at least one per-chamber mesh with an ontogenetic order, fit a sphere by least squares to the vertices of each chamber mesh and to the whole-test mesh (subsample up to 5000 vertices if a mesh is larger). The residual s is the root-mean-square of |distance to the fitted centre minus radius|, divided by the fitted radius. The specimen value is the median s of its chambers excluding the earliest chamber, minus the whole-test s; negative means the chambers are closer to a sphere than the whole shell. SUPPORTS if n_specimens >= 10, a majority of specimens have a negative value, and a one-sided Wilcoxon signed-rank test of the values against 0 has p < 0.05 in the negative direction. If chamber meshes are absent and the archive instead gives per-chamber surface area and volume, the same test is run on the isoperimetric quotient Q = 36*pi*V^2/A^3, with the sign reversed (chambers should have higher Q than the whole test). Meshes are preferred when both exist. The choice is made from which files the archive contains, not from the sign of the result." . "false"^^ . . "per-chamber STL meshes and the shrink-wrapped whole test for planktonic foraminifera, plus the measurement tables used to name species" . "CC BY 4.0" . . "Dataset: Micro-CT scans, whole-test meshes, and internal chamber segments of planktonic foraminifera" . . . . . . "sha256:93ab4296c45d813c3bbe2f5b71c674434a45a72d98feaf71db8820dc6502e996" . "sha256:a939b1412a4ae9060e608ec05d2f2f0a8b224010c7d86f397a4cc4ad47bb1f7e" . "sha256:c749ca03a41d0c7d8affec21c188818eb58f21837eb9eee94c757f985f37f2cb" . "sha256:daa5cb899b8346652d1ad3e75dbff9945abab677d9c950ef4eeede4ae796b683" . . "delphisci claim tester (grok-4.7-high-fast; cursor/thompson-form-claims-13cd)" . . . "grok-4.7-high-fast" . . . . . . . . . "2026-11-08T16:24:31.897Z"^^ . . . . . "Claim test of thompson1917-ch05-08: supports, confidence 0.35" . . "RSA" . "MIIBIjANBgkqhkiG9w0BAQEFAAOCAQ8AMIIBCgKCAQEAyFHtPB6p0mCjDozxlOY9CmO3nBm7Te3jFbXyLmM+G1tGe47Da0jYIbhkl9Oo1ZjxmCa3LCwlEqK8gj6hEIcL4ahW30JXh9HycfKVk1bugkzMEV15tJdTzAU3sGMFnbNsn43xY2LFLy45tApkRm5byEiM7oRRrooIdwoQKnP4Wzq/bfAoJ9qvW+MGkelyOTd1jbSt+KfmT7xKmDfWzAHddUlWwMX2QJOLxDaffOteTop6ZRgPiNmPJKA/S62Gac9XtSYT3oDkZ6/nr5cxxNLpOyxTf5FPISiCe11xYO/unYjY+mwnOfA8yhJIyx1l0O9I0vpOpn+E5TvyHJV+oSTrvwIDAQAB" . "o6RY+AQJTUkCyIJhbjsL572x8mc1pSrpTuIXhGtbGIXWp2bjCFnNSMWGZwnM6hlvqryp0fXQWt7OD8Ppluhq17fDa1hVX+YOrgewylFjM6L2RdEmuZq/UiH9ijJyW9lvQhM6oL330v8RoyM7Rlw85HFoI9dzJAhNyfREte2OAA3lsZjldqmAjc0pGZt/8HGt+19M6F1aoXvF5OvwXmW6oRv7FLTfBHpbTyAcTPgIURMTVsupInThbiooEIPJc+qwIdsU33AhUpTE0ZhnujgI4i+PmSX+cJjif5ooMwBI1HINUuveLCfVcpPM8WyhoNJmjZCJ32hjZndovRTUCcLc6Q==" . . .