# R1 — references, and how far each was checked

Checked 2026-11-14 by a web pass against Crossref DOI records, arXiv and library catalogues. **Bibliographic** means
the citation itself; **content** means the specific claim we lean on. Most publisher full texts were paywalled, so
"content not re-read" is common and is said each time. Nothing here has been checked against a physical copy.
Before submission every entry marked *partial* or *not checked* needs a page reference from someone holding the source.

## A. Strength of bundles and load sharing (what PULL's bench is a toy version of)

| | Source | Bibliographic | Content we use | Content status |
|---|---|---|---|---|
| A1 | H. E. Daniels (1945), "The statistical theory of the strength of bundles of threads. I", *Proc. R. Soc. Lond. A* 183(995), 405–435. doi:10.1098/rspa.1945.0011 | confirmed | an equal-load-sharing bundle's strength is asymptotically normal, mean per fibre max ₓ x[1 − F(x)] | not re-read |
| A2 | W. Weibull (1951), "A Statistical Distribution Function of Wide Applicability", *J. Appl. Mech.* 18(3), 293–297. doi:10.1115/1.4010337 | confirmed | the weakest-link strength distribution (our link strengths) | not re-read |
| A3 | B. D. Coleman (1958), "On the strength of classical fibres and fibre bundles", *J. Mech. Phys. Solids* 7(1), 60–70. doi:10.1016/0022-5096(58)90039-5 | confirmed | a loose bundle of Weibull fibres is weaker than the mean fibre, more so the wider the scatter (our "loose bundle 0.70") | not re-read |
| A4 | B. W. Rosen (1964), "Tensile failure of fibrous composites", *AIAA J.* 2(11), 1985–1991. doi:10.2514/3.2699 | confirmed | chain of bundles; a broken fibre carries load again beyond an ineffective length | not re-read |
| A5 | D. G. Harlow, S. L. Phoenix (1978), "The Chain-of-Bundles Probability Model for the Strength of Fibrous Materials I: Analysis and Conjectures", *J. Compos. Mater.* 12(2), 195–214, doi:10.1177/002199837801200207; "II: A Numerical Study of Convergence", 12(3), 314–334, doi:10.1177/002199837801200308 | confirmed | local load sharing | not re-read |
| A6 | H. L. Cox (1952), "The elasticity and strength of paper and other fibrous materials", *Br. J. Appl. Phys.* 3(3), 72–79. doi:10.1088/0508-3443/3/3/302 | confirmed | shear lag: load enters a fibre through interface shear over a length | not re-read |
| A7 | A. Kelly, W. R. Tyson (1965), "Tensile properties of fibre-reinforced metals: Copper/tungsten and copper/molybdenum", *J. Mech. Phys. Solids* 13(6), 329–350. doi:10.1016/0022-5096(65)90035-9 | confirmed | the critical transfer length l_c = σ_f d / 2τ (TWIST's "shortest fibre that holds") | formula from memory |

## B. Twist, helices and lateral pressure

| | Source | Bibliographic | Content we use | Content status |
|---|---|---|---|---|
| B1 | J. W. S. Hearle, P. Grosberg, S. Backer (1969), *Structural Mechanics of Fibers, Yarns, and Fabrics*, vol. 1, Wiley-Interscience, New York | confirmed (catalogues; no DOI) | twisted-yarn mechanics; lateral pressure from tension and curvature | **not checked**: the book could not be opened |
| B2 | Gégauff (1907), *Bull. Soc. Ind. Mulhouse* 77, 153 | **uncertain**: one reference list gives 177:150; original not seen, title unverified | filament strain = yarn strain × cos²θ, so yarn modulus ∝ cos²θ (our `laid`: stiffness cos³θ along the rope) | from secondary literature |
| B3 | M. M. Platt (1950), "Mechanics of Elastic Performance of Textile Materials. III. Some Aspects of Stress Analysis of Textile Structures—Continuous-Filament Yarns", *Text. Res. J.* 20(1), 1–15. doi:10.1177/004051755002000101 | confirmed | twist angle and yarn modulus, after Gégauff | from secondary literature |
| B4 | J. W. S. Hearle, H. M. A. E. El-Behery, V. M. Thakur (1959), "The mechanics of twisted yarns: tensile properties of continuous-filament yarns", *J. Text. Inst. Trans.* 50(1), T83–T111. doi:10.1080/19447025908662487 | confirmed | as B3 | not re-read |
| B5 | G. A. Costello, *Theory of Wire Rope*, Springer, 1990; 2nd ed. 1997. doi:10.1007/978-1-4612-1970-5 | confirmed | a helical wire's contact line load. **Convention:** Costello measures the helix angle α from the cross-section, so curvature is cos²α / r; our sin²θ / r is the same with θ from the axis. **The law:** X = N′τ − Tκ′, which is T × curvature only when the wire's bending shear N′ is neglected (ours has none: bead chains don't bend-resist) | checked in a thesis reproducing the framework, not in the book |
| B6 | J. W. Phillips, G. A. Costello (1973), "Contact Stresses in Twisted Wire Cables", *J. Eng. Mech. Div. ASCE* 99(2), 331–341. doi:10.1061/JMCEA3.0001739 | confirmed | as B5 | not re-read |
| B7 | I. M. Stuart (1961), "Capstan equation for strings with rigidity", *Br. J. Appl. Phys.* 12(10), 559–562, doi:10.1088/0508-3443/12/10/309; J. H. Jung, N. Pan, T. J. Kang (2008), "Capstan equation including bending rigidity and non-linear frictional behavior", *Mech. Mach. Theory* 43(6), 661–675, doi:10.1016/j.mechmachtheory.2007.06.002 | confirmed | normal load per length = tension / radius of curvature, and the capstan equation from it | not re-read; Euler's and Eytelwein's originals not checked |
| B8 | J. W. S. Hearle (1965), "Theoretical Analysis of the Mechanics of Twisted Staple Fiber Yarns", *Text. Res. J.* 35(12), 1060–1071, doi:10.1177/004051756503501202; J. Gregory (1953), "Cotton Yarn Structure Part IV", *J. Text. Inst. Trans.* 44(11), T499–T514, doi:10.1080/19447025308662613 | confirmed | staple yarn strength rises with twist (slippage) then falls (obliquity) | **the statement was not seen verbatim** in anything readable |

## C. Braids

| | Source | Bibliographic | Content we use | Content status |
|---|---|---|---|---|
| C1 | C.-P. Chou, B. Hannaford (1996), "Measurement and modeling of McKibben pneumatic artificial muscles", *IEEE Trans. Robot. Autom.* 12(1), 90–102. doi:10.1109/70.481753 | confirmed | a braided sleeve's force vanishes at a braid angle of 54.7° from the axis (factor 3cos²θ − 1) | formula from memory; the paper may not call the angle "neutral" |
| C2 | J. T. Evans, A. G. Gibson (2002), "Composite angle ply laminates and netting analysis", *Proc. R. Soc. A* 458(2028), 3079–3088. doi:10.1098/rspa.2002.1066 | confirmed | netting analysis: hoop to axial stress = tan²θ (our `trap` law's tan²θ) | not re-read |
| C3 | D. Brunnschweiler (1954), "The structure and tensile properties of braids", *J. Text. Inst. Trans.* 45(1), T55–T77. doi:10.1080/19447025408662631 | confirmed | braid geometry and jamming | not re-read |
| C4 | H. A. McKenna, J. W. S. Hearle, N. O'Hear (2004), *Handbook of Fibre Rope Technology*, Woodhead | confirmed | **nothing yet.** That a straight core is stiffer than its braided cover and loads first is our bench's result and plain mechanics (cos²θ); it was **not found** in any text that could be read, so it must not be attributed to this book without a page |  |
| — | "finger trap" as a term for a sleeve that tightens when pulled | **no citable source found** | use C1 and C2 for the mechanism and drop the nickname, or keep it as a nickname only |  |

## D. Shells, fits, packings (PULL's press: `js/skin.js`)

| | Source | Bibliographic | Content we use | Content status |
|---|---|---|---|---|
| D1 | S. Timoshenko, S. Woinowsky-Krieger, *Theory of Plates and Shells*, 2nd ed., McGraw-Hill, 1959 | confirmed (catalogues) | thin vessels: a cylinder's hoop stress PR/t, a sphere's PR/2t (the bench's `wall`) | location in the book from memory |
| D2 | S. P. Timoshenko, J. N. Goodier, *Theory of Elasticity*, 3rd ed., McGraw-Hill, 1970 | **not checked** | shrink fits and Lamé's solution (what "HOOP = half strength" means physically) |  |
| D3 | R. Schneider, *Convex Bodies: The Brunn–Minkowski Theory*, Cambridge University Press, 1993 (doi:10.1017/CBO9780511526282); 2nd expanded ed. 2014 | confirmed | Steiner's formula for a parallel body's area and mean curvature (the `HULL` estimate) | section not located |
| D4 | K. Schütte, B. L. van der Waerden, "Das Problem der dreizehn Kugeln", *Math. Ann.* 125(1), 325–334. doi:10.1007/BF01343127 (dated 1952 in the publisher's record; usually cited as 1953) | confirmed | twelve balls round one, and no more | — |
| D5 | J.-N. Roux (2000), "Geometric origin of mechanical properties of granular materials", *Phys. Rev. E* 61(6), 6802–6836, doi:10.1103/PhysRevE.61.6802; C. F. Moukarzel (1998), "Isostatic Phase Transition and Instability in Stiff Granular Materials", *Phys. Rev. Lett.* 81(8), 1634–1637, doi:10.1103/PhysRevLett.81.1634 | confirmed | **stated carefully:** a generic packing of rigid frictionless spheres is isostatic and its forces are fixed by geometry. A regular packing with more contacts than that, like ours (36 contacts among 13), is over-determined, and then the forces depend on the contacts' stiffness. That is why `sharing` gives a spread | abstracts read |

## E. Topology (CAD⁴'s braids)

| | Source | Bibliographic | Content we use | Content status |
|---|---|---|---|---|
| E1 | C. Damiani (2017), "A journey through loop braid groups", *Expo. Math.* 35(3), 252–285. doi:10.1016/j.exmath.2016.12.003; arXiv:1605.02323 | confirmed | the loop braid group as motions of unlinked circles in R³; its identification with ribbon braids in the 4-ball and with welded braids; the σᵢ generate a copy of the classical braid group (Remark 4.4) | **read** (arXiv text) |
| E2 | T. E. Brendle, A. Hatcher (2013), "Configuration spaces of rings and wickets", *Comment. Math. Helv.* 88(1), 131–162. doi:10.4171/CMH/280 | confirmed | the space of unlinked circles and its fundamental group | abstract |
| E3 | J. C. Baez, D. K. Wise, A. S. Crans (2007), "Exotic statistics for strings in 4d BF theory", *Adv. Theor. Math. Phys.* 11(5), 707–749. doi:10.4310/ATMP.2007.v11.n5.a1 | confirmed | the loop braid group's presentation (the braid-permutation group) | abstract |
| E4 | D. M. Dahm (1962), "A Generalization of Braid Theory", PhD thesis, Princeton (unpublished); D. L. Goldsmith (1981), "The theory of motion groups", *Michigan Math. J.* 28(1), 3–17. doi:10.1307/mmj/1029002454 | Goldsmith confirmed; Dahm via citing papers only | motion groups of circles | — |
| E5 | S. Satoh (2000), "Virtual knot presentation of ribbon torus-knots", *J. Knot Theory Ramifications* 9(4), 531–542. doi:10.1142/S0218216500000293 | confirmed | the **Tube map**: a (welded) braid gives tubes in 4-space. *Not* Artin's spinning, which turns a knotted arc into a knotted sphere | — |
| E6 | P. Bellingeri, A. Bodin (2016), "The braid group of a necklace", *Math. Z.* 283(3–4), 995–1010. doi:10.1007/s00209-016-1630-0; arXiv:1404.2511 | confirmed | "the natural map Bₙ → WBₙ sending σᵢ into σᵢ is injective": every classical braid is a distinct loop braid | **read** (arXiv text) |
| E7 | R. Fenn, R. Rimányi, C. Rourke (1997), "The braid-permutation group", *Topology* 36(1), 123–135. doi:10.1016/0040-9383(95)00072-0 | confirmed | as E3 | — |

**The gap in E, closed (2026-11-14).** Revolving a knot that lies in a half-space about the half-space's face is the
*spun torus* of the knot, and it is the same surface as Tube(K): B. Winter (2009), "The classification of spun torus
knots", *J. Knot Theory Ramifications* 18(9), 1287–1298, doi:10.1142/S0218216509007476, arXiv:0711.1638, Theorem 3.4
(**read**); Theorem 3.5, "Tube preserves the knot group and quandle" (**read**). So CAD⁴'s `tubeknot` is the spun
trefoil, its complement has the trefoil's group, and it is knotted (the step from "non-abelian group" to "knotted" is
standard, not quoted). The same for braids is by analogy: E1 and E6 give the embedding of the braid group; that the
revolved plait is its image has still not been seen stated for braids.

## F. Knots, one-sided surfaces, Borromean links, the belt trick (checked 2026-11-14)

| | Source | Bibliographic | Content we use | Content status |
|---|---|---|---|---|
| F1 | A. Skopenkov, "Embedding and knotting of manifolds in Euclidean spaces", arXiv:math/0604045 (published in *Surveys in Contemporary Mathematics*, LMS Lecture Note Ser. 347, 2008, 248–342: recalled, not checked) | arXiv confirmed | every embedded circle in R⁴ is unknotted: Theorem 2.1.a (any n-manifold unknots in Rᵐ for m ≥ 2n + 2) and Theorem 2.5 (spheres unknot in codimension ≥ 3). CAD⁴'s `knot` | **read** |
| F2 | E. C. Zeeman (1960), "Unknotting spheres", *Ann. of Math.* 72, 350–360, doi:10.2307/1970138; A. Haefliger (1961), "Plongements différentiables de variétés dans variétés", *Comment. Math. Helv.* 36, 47–82, doi:10.1007/BF02566892 | confirmed | the primary sources for F1. (Zeeman 1963, "Unknotting combinatorial balls", is about ball pairs: cite 1960 for spheres) | not read |
| F3 | B. Davis, H. Howards, J. Newman, J. Parsley (2011), "An infinite family of convex Brunnian links in Rⁿ", *Geom. Dedicata* 155, 141–149, doi:10.1007/s10711-011-9581-4, arXiv:1101.4863 | confirmed | Theorem 3.1: in R⁴ two round 2-spheres and an ellipse form a Brunnian link, with explicit equations (CAD⁴'s `borromean` is theirs scaled by 0.4). **Only this triple:** three circles never link in R⁴ (F1); for one sphere and two circles no source was found and a short argument says none exists (unsourced) | **read** |
| F4 | T. Yanagawa (1964), "Brunnian systems of 2-spheres in 4-space", *Osaka J. Math.* 1(2), 127–132. doi:10.18910/6553 | confirmed | Brunnian systems made of 2-spheres alone exist in 4-space, so "two spheres and a ring" is one construction and not the only mixture | **read by the auditor** (construction in §2 and its theorem); not by us |
| F5 | G. Matić, F. Öztürk, J. Reyes, A. Stipsicz, G. Urzúa, "An exotic 5RP² in the 4-sphere", arXiv:2312.03617 | arXiv confirmed; no journal reference found | the complement of the standard projective plane in S⁴ has fundamental group ℤ/2: a loop round it twice can be pulled free. Homology alone (Alexander duality: H₁ of the complement ≅ H² of the surface, ℤ/2 for any one-sided surface) is textbook and was not sourced. **Standard embedding only** | **read** |
| F6 | E. D. Bolker (1973), "The spinor spanner", *Amer. Math. Monthly* 80(9), 977–984, doi:10.1080/00029890.1973.11993429; M. Staley (2010), "Understanding quaternions and the Dirac belt trick", *Eur. J. Phys.* 31, 467–478, doi:10.1088/0143-0807/31/3/004 | confirmed | π₁(SO(3)) = ℤ/2 against π₁(SO(2)) = ℤ: TWIST's "a hose saves one turn" | not read |
| F7 | N. D. Mermin (1979), "The topological theory of defects in ordered media", *Rev. Mod. Phys.* 51(3), 591–648, doi:10.1103/RevModPhys.51.591 | confirmed | a body with rotational symmetry group G has configuration space SO(3)/G, whose fundamental group is G's lift to SU(2), twice as big. For our packed thirteen (the cuboctahedron's 24 rotations) that is 48 classes, so **some part turns are stubborn as well as the whole one**: TWIST's caveat is real | general rule recalled; the specific "48" is inference |
| F8 | J. C. Baez, D. K. Wise, A. S. Crans (2007), Theorem 2 (E3); C. Damiani (2017), Prop. 3.14 (E1) | confirmed | the loop braid group's presentation. Of the two mixed relations σσρ = ρσσ, **exactly one holds** (which one depends on the convention: Baez et al. and Damiani differ); the other is the "forbidden move". And a ring passing through its neighbour is a different element from the neighbour passing through it. Copy one source's relations verbatim | **read**, and checked by computing in Aut(F₃) |

| E8 | B. Winter (2009), "The classification of spun torus knots", *J. Knot Theory Ramifications* 18(9), 1287–1298. doi:10.1142/S0218216509007476; arXiv:0711.1638 | confirmed | revolving a knot in a half-space gives its spun torus, which is Tube(K) (Thm 3.4) and has the knot's group (Thm 3.5) | **read** |

## G. The setting, and the nearest physics (from PRIOR-ART.md; statuses as recorded there)

| | Source | Bibliographic | Content we use | Content status |
|---|---|---|---|---|
| G1 | Hqak (wxyhly), "4D Space (X): Knots and Links", 2019, wxyhly-en.github.io/archives/knot4d | page seen | chains of rings fall apart in 4D; rings link spheres; unequal tori link | page read by the prior-art pass |
| G2 | Hqak (wxyhly), "4D World (VI): Newtonian Mechanics", 2020, wxyhly-en.github.io/archives/newton4 | page seen | a rigid-body simulation of a linked chain in 4D | page read by the prior-art pass |
| G3 | Hqak (wxyhly), 2016, wxyhly-en.github.io/archives/lavie4ds | page seen | 4D cloth is woven of sheets; beads and shells on a string | page read by the prior-art pass |
| G4 | A. Seguin, J. Crassous (2022), *Phys. Rev. Lett.* 128, 078002 | record only | why twist grips in three dimensions: frictional amplification with twist | not read |
| G5 | P. B. Warren, R. C. Ball, R. E. Goldstein (2018), *Phys. Rev. Lett.* 120, 158001 | record only | why staple yarn holds: tension transmission through friction | not read |
| G6 | hi.gher.space forum threads "4d weaving" (2010) and "N-Dimensional Knots" | **could not be fetched** | may contain parts of sections 3.2 to 3.3 | unread |

| G7 | hi.gher.space forum, "N-Dimensional Knots" and "4D Knots" threads | **search summaries only** | "to make a knot in n-dimensional space, the knotting object must be (n−2)-dimensional"; "for actual weaving, one uses a cloth that is two dimensions less than all-space" | not read; the site and its archives were unreachable on 2026-11-16 (one thread has a Wayback snapshot: web.archive.org/web/20190720115541/http://hi.gher.space/forum/viewtopic.php?f=24&t=2029) |

G6, corrected 2026-11-16: the "4d weaving" thread (opened 28 October 2010) is known from search summaries only: strings
would fall off each other, so a 4D being would weave with sheets. G1, added the same day: the page also says textile
threads would be hollow tubes, which is the tube fibre of section 3.1; credit it.

## H. Twist kept modulo two (checked 2026-11-16)

| | Source | Bibliographic | Content we use | Content status |
|---|---|---|---|---|
| H1 | D. Kosanović, "On fundamental groups of spaces of framed embeddings of a circle in a 4-manifold", arXiv:2407.06923 | arXiv confirmed | §1.2: an embedded circle in a 4-manifold has two framings "because π₁ SO₃ ≅ ℤ/2", where a knot in a 3-manifold has ℤ-many | **read** |
| H2 | A. Putman, "Homotopy groups of spheres and low-dimensional topology", lecture notes, academicweb.nd.edu/~andyp/notes/HomotopySpheresLowDimTop.pdf | confirmed | Lemma 5.3: π₁(SOₙ) is ℤ for n = 2 and ℤ/2 for n ≥ 3; the ℤ/2 invariant of framed circles | **read** |
| H3 | E. Fadell, J. Van Buskirk (1962), "The braid groups of E² and S²", *Duke Math. J.* 29, 243–257 | **from memory; not checked** | in the braid group of the sphere the full twist has order two | not read. This is the statement section 3.1 leans on; check it in the original |
| H4 | G. H. M. van der Heijden, M. A. Peletier, R. Planqué, "A consistent treatment of link and writhe for open rods, and their relation to end rotation", arXiv:math-ph/0310057 | arXiv confirmed | for a clamped rod in three dimensions: "Rods with any even number of end turns (including zero) lie in one class; rods with any odd number of turns lie in the other", by the belt trick | **read** |

Not found, after about fifteen searches: any statement that a twisted rod, yarn or bundle in four dimensions keeps
twist only modulo two. Both ingredients are in print separately.

Page-level checks made 2026-11-16 (supersede the entries above where they differ): Costello's equations were read in a
thesis that reproduces them (Sura 1984, Liverpool: X = N′τ − Tκ′, helix angle from the cross-section); the vessel
formulas in Roylance's MIT notes (sphere pr/2b, cylinder pr/b) and in Timoshenko's 1940 first edition, p. 365; the
shrink fit in Timoshenko's *Strength of Materials* II (1930), art. 41; Chou and Hannaford's (3cos²θ − 1) and 54.7° in
the paper itself, which gives the angle no name; the Daniels value for Weibull fibres, (ρe)^(−1/ρ) / Γ(1 + 1/ρ), about
0.65 at modulus 5, from Porwal, Beyerlein and Phoenix (2006); Seguin and Crassous, and Warren, Ball and Goldstein, in
their arXiv versions (2110.04206; 1804.07606); the Brunnian link's equations in arXiv:1101.4863. Still unread:
Hearle, Grosberg and Backer; Evans and Gibson's own statement of tan²θ; Kelly and Tyson.

Not found: the Price–Roseman paper on embeddings of the projective plane as a journal article (it is cited as a 1975
University of Iowa preprint); any paper titled "Borromean rings in higher dimensions".

