Pith. sign in

REVIEW 3 major objections 4 minor 1 cited by

Almost toric fibrations on K3 surfaces via degenerations

T0 review · 3 major / 4 minor · reviewed 2026-08-08 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper proves that every smooth fiber of a symplectic type III Kulikov model of a K3 surface carries an almost toric fibration over the intersection complex.

desk verdict Real, original construction of ATFs on K3 generic fibers from symplectic Kulikov models; the Gross–Siebert comparison is stated too broadly and needs a transversality hypothesis. read the letter →

arxiv 2502.04304 v3 pith:GOVXJDQI submitted 2025-02-06 math.SG

Pranav Chakravarthy , Yoel Groman This is my paper · ORCID
classification math.SG MSC 14H7014J2814J3214D06
keywords K3surfacesalmosttoricfibrationssymplecticKulikovmodelsnodalintegralaffinestructuresSYZconjecturedegenerationssmallresolutions
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper establishes that the smooth fibers of a maximal degeneration of K3 surfaces always carry an almost toric fibration—a Lagrangian torus fibration whose base has a nodal integral affine structure with finitely many singular points. The base is the intersection complex built from the central fiber of a type III Kulikov model, and the affine structure on it is determined by the almost toric fibrations on the components of the central fiber. The paper further shows that smooth anticanonical hypersurfaces in smooth toric Fano threefolds admit such Kulikov models, and that the induced affine structure on the intersection complex agrees, up to nodal slides, with the nodal affine structure on the boundary of the moment polytope used in mirror symmetry. This gives a concrete SYZ picture in real dimension four: near the large complex structure limit, the K3 fiber is fibred by Lagrangian tori over the expected skeleton.

What carries the argument

The argument is carried by the intersection complex $\Delta_X$ equipped with a nodal integral affine structure, together with two mechanisms. The first is symplectic parallel transport from the central fiber, modified near triple intersection points so that it pulls the glued almost toric fibrations back to a continuous Lagrangian torus fibration on the smooth fiber. The second is an interpolation step over the remaining 1-skeleton intervals, where the preimage of a rectangular neighborhood is shown to be symplectomorphic to $R \times T^2$ using the classification of minimal strong fillings of the standard contact $T^3$ and the classification of symplectic forms on $S^2 \times S^2$. For the toric Fano case, the key additional object is the cycle $\Sigma_{\vec{t}} = \sum_i t_i \widetilde{\Sigma}_i$ formed from proper transforms of base-locus components; its positive intersection numbers with all exceptional curves are what allow the small resolution to carry a symplectic form taming the complex structure.

What would settle it

Compute the monodromy of the induced integral affine structure around a loop in the intersection complex of a concrete symplectic Kulikov model that encloses no node; if the monodromy is nontrivial, Theorem 2.10 and hence Theorem 1.6 would fail.

Watch

Extended reading notes

Core claim

Theorem 1.6 is the central claim: given a symplectic Kulikov model of type III, the generic smooth fiber $X_t$ with the symplectic form $\omega|_{X_t}$ admits an almost toric fibration $\mu: X_t \to \Delta_X$ over the intersection complex $\Delta_X$, inducing the nodal integral affine structure obtained in Lemma 1.5 by gluing the almost toric fibrations of the central-fiber components. The proof pulls the glued central-fiber fibration back to the smooth fiber by symplectic parallel transport, modifies the projection so the transport is smooth near triple points, and then extends the fibration across the remaining intervals of the 1-skeleton using classifications of symplectic fillings of the standard contact three-torus and of symplectic structures on $S^2 \times S^2$. The second main result, Theorem 1.8, constructs such a Kulikov model for a smooth anticanonical hypersurface in a smooth toric Fano threefold by resolving the 24 double points of the anti-canonical pencil through a carefully ordered sequence of small blow-ups, and proves that the induced affine structure is integral affine isomorphic, up to nodal slides, to the nodal integral affine structure on the boundary of the moment polytope.

Load-bearing premise

The load-bearing premise is that the anticanonical hypersurface intersects every positive-dimensional toric stratum transversely, so that the total space of the pencil has only ordinary double point singularities; the paper's own footnote says this transversality is established only near the tropical limit, although Theorem 1.8 is stated for every smooth anticanonical hypersurface.

Editorial extensions

If this is right

  • Every smooth fiber of a symplectic type III Kulikov model of a K3 surface carries a Lagrangian torus fibration over the intersection complex, realizing the SYZ picture for maximal degenerations in real dimension four.
  • For a smooth anticanonical hypersurface in a smooth toric Fano threefold, the restriction of any toric Kähler form admits an almost toric fibration whose base is, up to nodal slides, the boundary of the moment polytope with its standard nodal affine structure.
  • The affine structure on the base does not depend, up to nodal slides, on the order in which the double points are resolved or on the small positive parameters used to define the symplectic form on the resolution.
  • The intersection complex of the degeneration and the boundary of the moment polytope are identified as nodal integral affine manifolds, giving a direct bridge between the two bases that appear in the SYZ and toric-degeneration approaches to mirror symmetry.
  • This resolves, for K3 surfaces, the problem of producing a weak SYZ fibration on a Kähler K3 from a maximal degeneration whose total space is tamed by a symplectic form.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the preparatory step generalizes to higher-dimensional toric degenerations, as the authors expect, the hard step's reliance on four-dimensional classification results becomes the main obstruction; replacing it with a Lefschetz-fibration argument would be a natural testable extension.
  • The invariance of the affine structure up to nodal slides suggests that the almost toric base is a symplectic invariant of the K3 pair near the large complex structure limit; computing the node positions for two different degenerations of the same K3 would test this.
  • For anticanonical hypersurfaces, the theorem lets one read the SYZ base directly from the moment polytope data, so the locations of the nodal singularities of the fibration could be predicted computationally before any symplectic construction is carried out.
  • The transversality caveat in the proof suggests that the full statement of Theorem 1.8, for every smooth anticanonical hypersurface, may require a different argument than the one presented, perhaps a deformation to the tropical limit followed by a symplectic isotopy.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 4 minor

Summary. The paper studies symplectic degenerations of K3 surfaces. Theorem 1.6 asserts that for a type III symplectic Kulikov model, the generic smooth fiber carries an almost toric fibration over the intersection complex, with the nodal integral affine structure obtained from the ATFs on the central-fiber components. Theorem 1.8 claims that every smooth anticanonical hypersurface in a smooth toric Fano threefold, equipped with a restricted toric Kähler form, admits a symplectic Kulikov model whose induced integral affine structure is Gross–Siebert up to nodal slides. The proof combines a preparatory gluing/parallel-transport step, a hard step using Wendl's filling classification and classification results for symplectic log Calabi-Yau divisors, and a small-resolution construction with a greedy choice of blow-ups.

Significance. If the main construction is correct, Theorem 1.6 gives a concrete weak SYZ fibration for K3 degenerations of type III and connects the symplectic geometry of smooth fibers to the intersection complex and to Gross–Siebert structures. The explicit greedy small-resolution procedure in Section 4 and the comparison in Proposition 4.18 are valuable and checkable. The paper makes fair use of recent classification results [LMN22, LMN23, Wen10]. However, the stated generality of Theorem 1.8 exceeds what the proof establishes, and one step in the hard part is too compressed to be accepted as written.

major comments (3)
  1. [§4.1, Lemma 4.3 and §4.6, Theorem 1.8] Theorem 1.8 is stated for every smooth anticanonical hypersurface V, but Lemma 4.3 assumes that s1 intersects every positive-dimensional toric stratum transversely, and the footnote to Lemma 4.3 concedes that this transversality is known only when V is close to the tropical limit. Smoothness of V does not imply that condition: a smooth quartic in P^3 can have a tangency along a coordinate line, and in that case the local model z1 z2 = λ s1 with s1 ∈ m_p^2 gives a quadratic part of rank 2 rather than an ordinary double point. Consequently the small resolution of §4.1, the symplectic form construction of Lemma 4.4/Lemma 4.9, and the ATF construction of Lemma 4.13 are not available for such V. The proof therefore supports only the transversality-generic or tropical-limit version of Theorem 1.8; the theorem and the abstract should be restricted accordingly, or a separate argument for non-transverse smooth V must be supplied.
  2. [§3.1, Lemma 3.3] The minimality claim needed to apply Wendl's classification is dispatched in a single sentence: "For minimality it suffices to prove that μ_1^{-1}(R') is topologically aspherical. For this note the μ_1^{-1}(R) is contained in a normal neighborhood of a cylinder (the sphere with two little discs removed) in the total space of the fibration." The containment is not proved, and it is not immediate from Proposition 2.17 that the preimage of a thickening of an edge-interval is homotopy equivalent to an annulus or has contractible universal cover. Since the rest of the hard step depends on [Wen10, Corollary 4], this step needs a complete argument.
  3. [§3.3, Proposition 3.7] The proof of Proposition 3.7 reduces Θ_der to a Hamiltonian isotopy on a neighborhood of each component D_i and notes that the Hamiltonian can be taken to vanish near D_ij, but the final passage to a single Hamiltonian isotopy on the entire neighborhood of D, compatible with these overlaps, is not written out. The gluing is plausible by the stated vanishing near the vertices, but since Proposition 3.7 is the interpolation step that closes the proof of Theorem 1.6, the argument should be made explicit.
minor comments (4)
  1. [§2.5, Lemma 2.16] The Donaldson criterion is printed as "|∂π_r| < |∂π_r|"; this is self-comparing and should read "|∂π_r| < |\bar∂π_r|" (or the appropriate variant from [Don96]). The displayed inequalities in the same proof should be checked with the corrected notation.
  2. [§4.1, proof of Lemma 4.3] In the last sentence of the proof, "By Lemma 4.3 these are precisely..." should refer to Lemma 4.2; as written it cites the lemma being proved.
  3. [§4.4, after Lemma 4.13] Theorem 3.6 is repeatedly called the "Torelli Theorem"; it is a classification result for symplectic log Calabi-Yau divisors from [LMN22], not the classical Torelli theorem, and the label should be corrected.
  4. [§4.2, Lemma 4.12] In the proof of Lemma 4.12, "perturbations of the support constants defining C_i" should read "perturbations of the support constants defining the facets F_i", since C_i is the divisor associated to the facet.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the generic-fiber ATF is constructed from central-fiber ATFs via parallel transport plus external four-dimensional classification results, and the Gross-Siebert comparison is proved directly; the single self-citation [GV23] is not load-bearing.

full rationale

The derivation chain is not circular. Theorem 1.6 starts from a symplectic Kulikov model, which by Definition 1.2 already carries ATFs on the central fiber components, but the theorem's conclusion is a new ATF on the smooth fiber. The preparatory step (Sections 2.1-2.6) glues and modifies those ATFs, defines a nodal integral affine structure on the intersection complex, and uses symplectic parallel transport; the hard step (Section 3) extends the fibration across the remaining intervals in the 1-skeleton using Wendl's classification of strong fillings of T^3, McDuff's classification of symplectic forms on S^2 x S^2, and the external Torelli-type result [LMN22]. None of these steps sets the output equal to the input by construction; the final fibration is genuinely assembled rather than assumed. Theorem 1.8 likewise proves existence of a symplectic Kulikov model by small-resolving a toric degeneration, constructing the symplectic form via Lemma 4.4 with the cycle of Lemma 4.6, and building ATFs on the resolved boundary components in Lemma 4.13; the comparison with the Gross-Siebert integral affine structure is then proved by the explicit induction in Proposition 4.18 using developing maps. The parameters t_i are arbitrary small decreasing constants, not fitted to any target quantity, and the claimed independence up to nodal slides is argued rather than postulated. The only self-citation, [GV23], is used for the standard Maslov-zero fact in Section 1.3 and is not load-bearing for either main theorem; it is also an externally published result rather than an author-supplied uniqueness theorem. There is, however, a non-circular correctness caveat worth flagging: Lemma 4.3 assumes the anticanonical section s1 intersects every positive-dimensional toric stratum transversely, and the footnote states that this holds only when V is close to the tropical limit, while Theorem 1.8 is stated for every smooth anticanonical hypersurface. This is a proof-generality gap, not a circular reduction, so it does not affect the circularity score.

Assumptions & free parameters 1 free parameters · 8 assumptions · 0 invented entities

The paper introduces no new free parameters that affect the final result, since the t_i cancel up to nodal slides. It relies on a chain of external classification theorems and on the input ATFs on the central fiber components. No new physical or mathematical entities are postulated.

free parameters (1)
  • greedy blow-up parameters t_1 > ... > t_m = arbitrary strictly decreasing positive reals
    Introduced in Lemma 4.6 to weight the base locus cycles when building the symplectic form on the small resolution. The final integral affine structure is independent of these values up to nodal slides, as shown in Proposition 4.18.
assumptions (8)
  • domain assumption Kulikov-Persson-Pinkham: every degeneration of K3 surfaces extends to a Kulikov model with trivial canonical bundle and reduced normal crossings central fiber.
    Invoked in the introduction to frame symplectic Kulikov models for K3 degenerations.
  • domain assumption LMN23: existence of almost toric fibrations on symplectic log Calabi-Yau pairs (X_i, sum_j X_ij) with prescribed boundary cycle.
    Input defining a symplectic Kulikov model in Definition 1.2 and Remark 1.3.
  • domain assumption Wendl's classification (Corollary 4 of [Wen10]): every minimal strong filling of (T^3, xi_std) is diffeomorphic to T^2 x D^2.
    Used in Lemma 3.1 to extend the torus fibration across the interval A.
  • domain assumption McDuff-Lalonde classification of symplectic forms on S^2 x S^2 (Theorem 9.4.7 in [MS17]).
    Used in Proposition 3.4 to standardize the symplectic form after compactification.
  • domain assumption LMN22 Proposition 2.11: ambient symplectomorphism of omega-orthogonal log Calabi-Yau divisors with matching componentwise homology classes.
    Used in Proposition 3.4 and Lemma 4.13 to identify divisors and pull back almost toric fibrations.
  • domain assumption STY02: existence of a symplectic form taming the small resolution when a 4-cycle has positive intersection with all exceptional spheres.
    Basis for the symplectic structure on the resolved total space in Section 4.2.
  • domain assumption FM13 Proposition 2.4: pi_0(Diff_id,x,y(S^2)) = Z generated by Dehn twists.
    Used in Proposition 3.11 to analyze symplectomorphisms of the fiber spheres in the interpolation step.
  • domain assumption Mik04 section 6.6: transversality of H with toric strata holds near the tropical limit.
    Footnote to Lemma 4.3; used to justify the genericity hypothesis for the pencil.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Almost toric fibrations on K3 surfaces via degenerations." pith.science (2026). https://pith.science/paper/GOVXJDQI

@misc{pith2026250204304,
  author       = {Pith},
  title        = {Pith review of: Almost toric fibrations on K3 surfaces via degenerations},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GOVXJDQI}},
  note         = {Machine review of arXiv:2502.04304}
}
read the original abstract

For K\"ahler K3 surfaces we consider Kulikov models of type III tamed by a symplectic form. Our main result shows that the generic smooth fiber admits an almost toric fibration over the intersection complex, which inherits a natural nodal integral affine structure from almost toric fibrations of the boundary divisors. We prove that a smooth anti-canonical hypersurface in a smooth toric Fano threefold, equipped with a toric K\"ahler form, admits a symplectic Kulikov model. Moreover, we demonstrate that the induced integral affine structure on the intersection complex is integral affine isomorphic (up to nodal slides) nodal integral affine structure considered by Gross and Siebert on the boundary of the moment polytope.

Figures

Figures reproduced from arXiv: 2502.04304 by the authors.

Figure 1
Figure 1. Monodromy around γ Theorem 2.10. There is a unique extension of the piecewise nodal integral affine structure on the intersection complex introduced in Proposition 2.4 to a strict nodal integral affine structure which agrees with the canonical structure of Definition 2.5 on neighborhoods of the vertices. Remark 2.11. The crux of the statement amounts to triviality of the monodromy of the integral affine structure on… view at source ↗
Figure 2
Figure 2. Local model of the resolution at a double point p defined by zizj = λs1. Remark 4.7. Note that the modification to the symplectic form is done by adding the Poincar´e dual of the base locus cycle which doesn’t meet the codimension 2 strata of the singular fibre. Therefore it can be taken to be supported away from the codi￾mension 2 strata in the blow up. As a result we have that the proper transforms of the codimens… view at source ↗
Figure 3
Figure 3. Almost toric blow-up 4.4.2. Construction of Almost toric fibration. We recall the following useful claims. Theorem 4.11. (Theorem 6.3 in [Gui94]) Let (M, ω, µ, ∆) be a symplectic Toric manifold with image of momentum map µ : M → R n being the Delzant polytope ∆. Let ui denote the primitive normal vector to the boundary facets and let λi denote the support constants defining the boundary hyperplanes. That is, ∆ is de… view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: The inclusion Q0 \ ∂1Q(0) → Q(1) \ ∂1Q(1) i ν1 ν1 Let Q(k) be the boundary of the polytope obtained from Pω by replacing the first k defining inequalities hνi , xi ≤ bi with hνi , xi ≤ bi + ti . We inductively construct piecewise integral affine isomorphisms fk : Q (k)…
Figure 5
Figure 5. Figure 5: Hybrid degeneration after performing small blow up on the face F1 For the base case start by blowing up the double points mapping to edges of F1 by an amount t1. To construct the map f1 we start with the obvious inclusion i : Q(0) \ ∂1Q(0) ֒→ Q(1) \ ∂1Q(1), illustrated…
Figure 6
Figure 6. Figure 6: Construction of f1 f1 [PITH_FULL_IMAGE:figures/full_fig_p036_6.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. The spaces of K\"ahler and holomorphically tamed symplectic forms on closed 4-manifolds

    math.SG 2026-07 accept novelty 7.0 of 10

    On closed 4-manifolds with b+2=1, Kodaira dimension zero, or negative curvature, holomorphically tamed symplectic forms with Kähler-type classes are Kähler, and Kähler-type forms in a fixed class are unique up to homo...

Reference graph

Works this paper leans on

29 extracted references · 26 canonical work pages · cited by 1 Pith paper

  1. [1]

    Javier Fern \'a ndez de Bobadilla and Tomasz Pe ka, Fibrations by lagrangian tori for maximal calabi-yau degenerations and beyond, arXiv preprint arXiv:2312.13248 (2023)

    work page Pith review arXiv 2023

  2. [2]

    S. K. Donaldson, Symplectic submanifolds and almost-complex geometry, Journal of Differential Geometry 44 (1996), no. 4, 666 -- 705

    work page 1996

  3. [3]

    3, 467--495

    Philip Engel, Looijenga’s conjecture via integral-affine geometry, Journal of Differential Geometry 109 (2018), no. 3, 467--495

    work page 2018

  4. [4]

    Jonny Evans, Lectures on lagrangian torus fibrations, London Mathematical Society Student Texts, Cambridge University Press, 2023

    work page 2023

  5. [5]

    Benson Farb and Danielle N Margalit, A primer on mapping class groups, 2013

    work page 2013

  6. [6]

    Mark Gross, Mirror symmetry and the strominger-yau-zaslow conjecture, arXiv preprint arXiv:1212.4220 (2012)

    arXiv 2012

  7. [7]

    Mark Gross and Bernd Siebert, Affine manifolds, log structures, and mirror symmetry, Turkish J. Math. 27 (2003), no. 1, 33--60. 1975331

    work page 2003

  8. [8]

    , Mirror symmetry via logarithmic degeneration data. I , J. Differential Geom. 72 (2006), no. 2, 169--338

    work page 2006

Show all 29 references
  1. [9]

    Algebraic Geom

    , Mirror symmetry via logarithmic degeneration data, II , J. Algebraic Geom. 19 (2010), no. 4, 679--780

    2010

  2. [10]

    2, 285 -- 309

    Victor Guillemin, Kaehler structures on toric varieties, Journal of Differential Geometry 40 (1994), no. 2, 285 -- 309

    1994

  3. [11]

    12, 2551--2637

    Yoel Groman and Umut Varolgunes, Locality of relative symplectic cohomology for complete embeddings, Compositio Mathematica 159 (2023), no. 12, 2551--2637

    2023

  4. [12]

    158, Cambridge University Press, Cambridge, 2016

    Daniel Huybrechts, Lectures on K3 surfaces , Cambridge Studies in Advanced Mathematics, vol. 158, Cambridge University Press, Cambridge, 2016

    2016

  5. [13]

    203--263

    Maxim Kontsevich and Yan Soibelman, Homological mirror symmetry and torus fibrations, Symplectic geometry and mirror symmetry, World Scientific, 2001, pp. 203--263

    2001

  6. [14]

    , Affine structures and non- A rchimedean analytic spaces , Springer, 2006

    2006

  7. [15]

    Vik. S. Kulikov, Degenerations of K3 surfaces and E nriques surfaces , Izv. Akad. Nauk SSSR Ser. Mat. 41 (1977), no. 5, 1008--1042

    1977

  8. [16]

    17, 2230009

    Yang Li, Survey on the metric SYZ conjecture and non-archimedean geometry , International Journal of Modern Physics A 37 (2022), no. 17, 2230009

    2022

  9. [17]

    Tian-Jun Li , Jie Min , and Shengzhen Ning , Enumerative aspect of symplectic log Calabi-Yau divisors and almost toric fibrations , arXiv e-prints (2022), arXiv:2203.08544

    2022 arXiv

  10. [18]

    , Almost toric presentations of symplectic log Calabi-Yau pairs , arXiv e-prints (2023), arXiv:2303.09964

    2023 arXiv

  11. [19]

    Michor, Topics in differential geometry, Graduate Studies in Mathematics, vol

    Peter W. Michor, Topics in differential geometry, Graduate Studies in Mathematics, vol. 93, American Mathematical Society, Providence, RI, 2008. 2428390

    2008

  12. [20]

    5, 1035--1065

    Grigory Mikhalkin, Decomposition into pairs-of-pants for complex algebraic hypersurfaces, Topology 43 (2004), no. 5, 1035--1065

    2004

  13. [21]

    Cheuk Yu Mak and Helge Ruddat, Tropically constructed L agrangians in mirror quintic threefolds , e58

  14. [22]

    Dusa McDuff and Dietmar Salamon, Introduction to symplectic topology, third ed., Oxford Graduate Texts in Mathematics, Oxford University Press, Oxford, 2017. 3674984

    2017

  15. [23]

    5, 953--972

    Johannes Nicaise, Chenyang Xu, and Tony Yue Yu, The non-archimedean SYZ fibration , Compositio Mathematica 155 (2019), no. 5, 953--972

    2019

  16. [24]

    Ulf Persson and Henry Pinkham, Degeneration of surfaces with trivial canonical bundle, Ann. of Math. 113 (1981), no. 1, 45--66

    1981

  17. [25]

    Wei-Dong Ruan, Lagrangian torus fibration of quintic hypersurfaces I: Fermat quintic case , AMS/IP Stud. Adv. Math. 23 (2001), 297--332

    2001

  18. [26]

    Smith, R.P

    I. Smith, R.P. Thomas, and S.-T. Yau, Symplectic conifold transitions, Journal of Differential Geometry 62 (2002), no. 2, 209 -- 242

    2002

  19. [27]

    Margaret Symington, Four dimensions from two in symplectic topology, Topology and geometry of manifolds (editors G Matić, C McCrory), Proc. Sympos. Pure Math. 71, Amer. Math. Soc. (2003) 153 MR2024634 (2003)

    2003

  20. [28]

    1-2, 243--259

    Andrew Strominger, Shing-Tung Yau, and Eric Zaslow, Mirror symmetry is T -duality , Nuclear Physics B 479 (1996), no. 1-2, 243--259

    1996

  21. [29]

    3, 337 -- 384

    Chris Wendl, Strongly fillable contact manifolds and J -holomorphic foliations , Duke Mathematical Journal 151 (2010), no. 3, 337 -- 384

    2010

Pith tools

Reviewed August 8, 2026 · model on record in the stance chip above.