Pith. sign in

REVIEW 5 minor 17 references

On ideal lemniscates, butterflies, and circular waves

T0 review · 0 major / 5 minor · reviewed 2026-08-27 · deepseek-v4-flash

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

Pith's one-line read Adding a length penalty to the ideal energy creates infinitely many lemniscate- and butterfly-shaped critical curves, plus non-closed circular waves.

desk verdict A genuinely new existence result for closed Jλ critical points, with a sound and detailed proof; deserves a serious referee. read the letter →

arxiv 2608.22396 v1 pith:JX3FFHJR submitted 2026-08-23 math.DG math.CA

Shinya Okabe , Glen Wheeler This is my paper · ORCID
classification math.DGmath.CA
keywords length-penalisedidealenergycurveslemniscatesbutterflycircularwavesFresnelphasesturning-windowminimisationstationaryimmersions
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

This paper proves that the length-penalised ideal energy $J_\lambda(\gamma)=\frac12\int_\gamma k_s^2\,ds+\lambda L[\gamma]$ has a much richer critical-point set than the free ideal energy $J_0$. For every $\lambda>0$ it constructs infinitely many geometrically distinct smooth closed immersed critical points with turning number zero, shaped like lemniscates and butterflies, together with infinitely many non-closed 'circular wave' stationary immersions whose tangent is periodic but whose position drifts each period. The lemniscate and butterfly branches are produced by a staged direct-minimisation scheme over intervals of lifted turning angle; a boundary-layer stationary-phase analysis shows that only two Fresnel phases, $3\pi/4$ and $7\pi/4$ modulo $2\pi$, allow the constituent quarter-arcs to glue smoothly. The circular waves bifurcate from multiply-covered semicircles at $\lambda=0$ via an implicit-function-theorem shooting argument, and each branch carries an intrinsic mean turning rate that separates the curves by index.

What carries the argument

The central object is the turning-window class $A^{(2)}(I)$ of doubled tangent-angle profiles, symmetric about the midpoint, with the closure constraint $\int_0^1\cos(\Phi H(t)+h(t))\,dt=0$ for the model profile $H(t)=\frac32 t-\frac12 t^3$. The argument turns on the endpoint stationary-phase estimate $\int_0^1\cos(\Phi H(t))\,dt=\sqrt{\frac{\pi}{6\Phi}}\cos(\Phi-\frac{\pi}{4})+o(\Phi^{-1/2})$; at the two phases where $\cos(\Phi-\frac{\pi}{4})=0$, a boundary-layer correction of energy $o(\Phi^{3/2})$ can restore closure at cost $o(\Phi^{3/2})$, making the window centre strictly lower in energy than its endpoints. That strict inequality forces the window minimiser's selected turning into the interior, which in turn forces the midpoint third-derivative jump $\theta_{sss}(\ell^-)=0$ and hence smooth gluing. For the waves, the machinery is the curvature-jet system $(k,p,q,\theta)$ with conserved constant $c=ab+\lambda$ and the nondegenerate Jacobian determinant $\ell^3/((2N+1)\pi)$ at the multiply-covered semicircle, which gives the implicit-function-theorem branches and the quadratic emergence of $\lambda$.

What would settle it

For a sequence $\Phi_n=2\pi n+\rho$ with $\rho$ near $3\pi/4$, compute the actual closure defect $\int_0^1\cos(\Phi_n H(t)+h_n(t))\,dt$ using an optimised correction $h_n$ with $\int(h_n'')^2=o(\Phi_n^{3/2})$; the theorem predicts that $\sqrt{\Phi_n}$ times the defect converges to $\sqrt{\pi/6}\cos(\rho-\pi/4)$. If the limit is not this Fresnel value, or if the midpoint balance defect $Q(\Phi)$ has no simple zero near the predicted phases for large $n$, then the good-window construction collapses and the smooth closed critical points may not exist.

Watch

Extended reading notes

Core claim

The central discovery is that adding a length penalty $\lambda>0$ to the ideal energy turns the $\lambda=0$ degenerate family of multiply-covered circles into two new infinite families of closed critical curves and one family of non-closed periodic-up-to-translation waves. For each $\lambda>0$ and each sufficiently large integer $n$, minimising $J_\lambda$ over the turning window $[2\pi n+\rho_j-\delta,2\pi n+\rho_j+\delta]$, $\rho_1=3\pi/4$, $\rho_2=7\pi/4$, yields a minimiser whose midpoint third-derivative jump vanishes, so four congruent quarter-arcs glue smoothly and close under x-axis reflection. The selected turning $\Phi^\dagger_{n,j}$ lies in the window interior, and the energy grows like $\lambda^{3/4}|\Phi^\dagger|^{1/2}$, so the curves are pairwise non-congruent. Separately, for each $N\in\mathbb{N}_0$, a shooting map at the multiply-covered semicircle base solution with $q(\ell)=\varepsilon$ produces a smooth branch with $\lambda_N(\varepsilon)=\frac{5}{4a_{N,*}^2}\varepsilon^2+o(\varepsilon^2)$, whose reflected and tiled extensions are stationary waves with mean turning rate $(2N+1)\pi/\ell$; after scaling, every $\lambda_*>0$ admits infinitely many such non-congruent waves.

Load-bearing premise

The load-bearing premise is the boundary-layer stationary-phase estimate that, for a quarter-arc with turning $\Phi$, the closure defect has leading term $\sqrt{\pi/(6\Phi)}\cos(\Phi-\pi/4)$ whenever the correction has bending energy $o(\Phi^{3/2})$; if that estimate failed at the phases $3\pi/4$ and $7\pi/4$, the window centre would no longer beat its endpoints, the minimiser's turning could sit at an endpoint, and the midpoint transversality forcing smooth closure would be lost.

Editorial extensions

If this is right

  • For every $\lambda>0$ there are infinitely many geometrically distinct smooth closed critical points of $J_\lambda$ with turning number zero, and their energy diverges as the index $n$ grows.
  • The two families are distinguished by the selected quarter-arc turning lying near $2\pi n+3\pi/4$ (lemniscates) or $2\pi n+7\pi/4$ (butterflies).
  • For every prescribed $\lambda_*>0$ there are infinitely many pairwise non-congruent, non-closed circular waves stationary for compactly supported variations.
  • Each circular wave has a well-defined mean turning rate $\Omega=(2N+1)\pi/\ell$, a congruence invariant, so waves from different indices $N$ cannot be congruent.
  • Closed super-lemniscates cannot be $J_\lambda$ critical points for $\lambda>0$; the only overlap with the $k_{ss}+c k^3=0$ family occurs in the open-curve setting at the special value $c=1/8$.

Reading between the lines

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

  • If the least-energy lemniscate critical points converge, after rescaling, to the conjectured homothetic 'ideal lemniscate' solution of the free ideal flow, then the whole countable family would provide infinitely many distinct homothetic solutions, linking the $\lambda>0$ and $\lambda=0$ pictures.
  • The same window-selection mechanism—make the centre of a turning window cheap through a stationary-phase cancellation—could transfer to other length-penalised curvature energies; a numerical scan of midpoint balance defects would show whether simple zeros occur only at the expected Fresnel phases.
  • The quadratic emergence $\lambda_N(\varepsilon)=\frac{5}{4a_{N,*}^2}\varepsilon^2+o(\varepsilon^2)$ indicates a pitchfork-type bifurcation from the multiply-covered circle; measuring the drift vector $P_{N,\varepsilon}$ near $\varepsilon=0$ would give a direct numerical signature of the symmetry-breaking.
  • The lower bound $J_\lambda[\Gamma]\ge\frac{16}{3}\left(\frac{9}{2}\right)^{1/4}\lambda^{3/4}|\Phi|^{1/2}$ may be approximately sharp in the large-$\Phi$ limit; comparing numerical energy ratios to this bound would test whether the constructed critical points are near-optimal among all closed curves.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Request a human review

A listed scientist reviews the paper for a fee and the review publishes here regardless of verdict. See the reviewers or get listed.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

0 major / 5 minor

Summary. The paper studies critical points of the length-penalised ideal energy Jλ on planar unit-speed immersions. The main result (Theorem 1.1) asserts that for every λ>0 there are infinitely many geometrically distinct smooth closed immersed critical points Γ_{n,j} with turning number zero, obtained from four congruent copies of a minimising quarter-arc whose selected turning lies near 2πn+3π/4 (lemniscates) or 2πn+7π/4 (butterflies), with energies tending to infinity. The construction minimises a doubled functional over turning windows; a boundary-layer stationary-phase analysis of the closure defect (Lemmas 4.2 and 4.3) creates 'good windows' whose interior minimisers satisfy the midpoint transversality condition θ_sss(ℓ^-)=0, allowing smooth closure by reflection. Theorem 1.3 and Corollary 5.15 construct, for each N, a one-parameter IFT branch of 'circular waves' bifurcating from multiply-covered semicircles at λ=0, giving non-closed stationary immersions with periodic tangent and nonzero translation per period; after scaling, this yields infinitely many pairwise non-congruent circular waves for every prescribed λ*>0. The paper also contains a Hamiltonian upgrade from the reduced angle ODE to the full geometric Euler-Lagrange equation (Proposition 4.9) and an appendix on the relation to super-lemniscates.

Significance. If correct, the results provide the first systematic existence theory for critical points of the singularly perturbed ideal energy in the closed setting beyond the multiply-covered circles, and identify a mechanism (boundary-layer selection of Fresnel phases) that is likely to be useful elsewhere. The paper's strengths are the concreteness of the construction: the sharp turning inequality is proved with its equality case, the closure-defect asymptotic is stated with explicit constants, the Jacobian in Lemma 5.3 is computed exactly, λ_N(ε) is derived from the ODE at second order rather than fitted, and the distinctness arguments use intrinsic invariants (energy growth, mean turning rate). The numerical figures are explicitly labelled illustrative and not used in any proof. The proof chain is internally consistent; I found no load-bearing error.

minor comments (5)
  1. [§4.2 (Lemma 4.2)] The proof of the endpoint stationary-phase estimate is compressed: the tail bound after cutting at y=R is asserted rather than displayed, although the estimate is standard and checkable. Since this lemma is load-bearing for the good-window construction, I recommend expanding the integration-by-parts step with the explicit control of Ψ'' and |g'_m|.
  2. [§5.8 (Corollary 5.15)] The recursive choice of ε_N is slightly implicit: the proof first says 'choose 0<|ε_N|<ε_{0,N}' and then observes that Ω̃_N tends to infinity as ε_N→0. It would be clearer to state that for each N the value Ω̃_N can be made arbitrarily large, and then choose the parameters recursively.
  3. [Appendix C.1] The numerical diagnostic identity M'_λ(Φ)=-2Q(Φ) is stated without derivation. Since the numerical content is explicitly illustrative this is not a correctness issue, but a one-line derivation (or a reference to the corresponding first-variation computation) would make the appendix self-contained.
  4. [Notation] The symbol N_0 is used for a threshold in Theorem 1.1 and N∈N_0 for the winding index in Theorem 1.3; the double use of 'N_0/N' can confuse a reader. A different symbol for the threshold (for example, N_*) would improve readability.
  5. [§2.7 (Proposition 2.13)] The statement of Proposition 2.13 says 'obtained from a unit-speed fundamental arc' without spelling out that the closed curve consists of four congruent copies; the proof uses four copies through Lemma 2.12. The statement should make the four-copy structure explicit.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the constructions are self-contained and no prediction reduces to a fitted input or self-citation.

full rationale

The paper's derivation chain is self-contained. The special phases 3π/4 and 7π/4 are derived from the stationary-phase expansion of the closure defect in Lemma 4.2, not read off from numerics or imposed as ansatz: the leading Fresnel term √(π/(6Φ)) cos(ρ−π/4) vanishes exactly at those phases, and Proposition 4.3 shows stability under corrections with bending energy o(Φ^{3/2}). The good-window inequality (49) is a theorem (Theorem 4.6) comparing the centre value, from Lemma 4.4, with endpoint values, from Lemma 4.5, using a strictly positive margin cΦ^{3/2}; it is not an empirical selection rule. The midpoint transversality θsss(ℓ⁻)=0 is forced by interiority of the selected turning (Proposition 4.8), and the upgrade from the reduced angle equation to the full geometric Euler-Lagrange equation is achieved by the Hamiltonian identity Λ=λ proven from length stationarity in Proposition 4.9, not by invoking an external uniqueness theorem. For the circular waves, the branch is found by the implicit function theorem with an explicitly computed nondegenerate Jacobian (Lemma 5.3), λ_N(ε) is computed by Taylor expansion along the ODE branch (Lemma 5.5), and non-closure comes from the first-order drift of the midpoint height (Lemma 5.9). Scaling to prescribed λ* is a direct dilation argument. The self-citations to [AMWW20], [MW26b], and [AW26] are used for context, comparison, or external classification results, and none is load-bearing for the main theorems; the cited uniqueness/classification result in [AMWW20] is an independent external theorem, and [MW26b] is used only in Appendix A to clarify that super-lemniscates are not the same curves. The numerical gallery is explicitly labelled 'purely illustrative and not used in any proof' (Appendix C), so the figures do not constitute fitted inputs renamed as predictions. No equation in the proof is equivalent to its own input by construction. The finding is therefore a normal non-finding: score 0.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

The proofs are self-contained: no numerical fitting, no empirically chosen constants. The only input parameters are λ>0 and the geometric class of curves; the special phases and branch families are derived from first principles.

assumptions (5)
  • standard math Sobolev embedding H^2(0,1) into C^1 and compactness of the inclusion are used to pass limits in the direct method (Lemma 3.3).
    Invoked repeatedly to obtain C^1 convergence of the rescaled angle from a uniform H^2 bound.
  • standard math The implicit function theorem for the shooting map (Theorem 5.4) is used to produce the circular-wave branches.
    The Jacobian determinant at the base point is computed and shown nonzero in Lemma 5.3.
  • standard math The stationary-phase expansion for oscillatory integrals (Lemma 4.2) is used to identify the good turning windows.
    This is a classical estimate; the proof gives the relevant tail bounds.
  • domain assumption The energy functional is defined for unit-speed immersions with curvature in H^1; the paper restricts to this class.
    The ideal energy requires square-integrable curvature derivative; all constructed curves are smooth, so the assumption is satisfied.
  • domain assumption The closure constraint ∫ sin θ ds = 0 is regular for minimisers because cos θ is not identically zero.
    Used to apply the Lagrange multiplier rule in Lemmas 2.7 and 4.1; proved via the constraint qualification in Lemma 2.7.

how reviews work

0 comments
Cite this review

Pith. "Pith review of On ideal lemniscates, butterflies, and circular waves." pith.science (2026). https://pith.science/paper/JX3FFHJR

@misc{pith2026260822396,
  author       = {Pith},
  title        = {Pith review of: On ideal lemniscates, butterflies, and circular waves},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JX3FFHJR}},
  note         = {Machine review of arXiv:2608.22396}
}
abstract

We prove the existence of infinitely many lemniscate-like, butterfly-like, and circular-wave critical points for the length-penalised ideal energy. For the lemniscate and butterfly family, we use a staged direct minimisation process to establish existence. A boundary-layer analysis reveals critical points near two Fresnel phases: $3\pi/4$ modulo $2\pi$, corresponding to lemniscates, and $7\pi/4$ modulo $2\pi$, which are the butterflies. The family of circular waves bifurcates, in a sense, from multiply-covered circles. We use an adapted shooting method to establish their existence.

Figures

Figures reproduced from arXiv: 2608.22396 by the authors.

Figure 1
Figure 1. The two least-energy numerically selected critical point candidates for [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Circular wave solutions for indices N = 0, 1, 2. For each N we shoot a semicircle-like stationary arc on [0, ℓ] with lifted endpoint angle θ(ℓ) = π 2 + (2N + 1)π and seam conditions ks(0) = ks(ℓ) = 0, then extend to a shooting cell [0, 2ℓ] by reflection and tile by the resulting translation vector. Each wave is then rescaled by the dilation Γ 7→ ρ Γ with ρ = λ(ε) 1/4 so that the stationary parameter becomes λ = 1. T… view at source ↗
Figure 3
Figure 3. Selected lemniscate-type critical point candidates for [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Selected butterfly-type critical point candidates for [PITH_FULL_IMAGE:figures/full_fig_p009_4.png]
Figure 5
Figure 5. Figure 5: Illustration of the “equality-case” fundamental arc obtained by integrating the curvature profile [PITH_FULL_IMAGE:figures/full_fig_p012_5.png]
Figure 6
Figure 6. Figure 6: The open curve obtained by integrating the tangent field [PITH_FULL_IMAGE:figures/full_fig_p039_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

17 extracted references · 17 canonical work pages

  1. [1]

    Closed ideal planar curves

    Ben Andrews, James McCoy, Glen Wheeler, and Valentina-Mira Wheeler. Closed ideal planar curves. Geometry & Topology , 24(2):1019--1049, September 2020

    work page 2020

  2. [2]

    Jellyfish exist

    Ben Andrews and Glen Wheeler. Jellyfish exist. arXiv:2601.21227, January 2026

    work page arXiv 2026

  3. [3]

    3D Euler spirals for 3D curve completion

    Gur Harary and Ayellet Tal. 3D Euler spirals for 3D curve completion. In Proceedings of the Twenty-Sixth Annual Symposium on Computational Geometry , SoCG '10, pages 393--402, New York, NY, USA, June 2010. Association for Computing Machinery

    work page 2010

  4. [4]

    The Euler spiral: a mathematical history

    Raph Levien. The Euler spiral: a mathematical history. Technical Report UCB/EECS-2008-111, EECS Department, University of California, Berkeley, September 2008

    work page 2008

  5. [5]

    From Spiral to Spline: Optimal Techniques in Interactive Curve Design

    Raphael Linus Levien. From Spiral to Spline: Optimal Techniques in Interactive Curve Design . PhD thesis, EECS Department, University of California, Berkeley, December 2009

    work page 2009

  6. [6]

    Minimum Curvature Variation Curves, Networks, and Surfaces for Fair Free-Form Shape Design

    Henry Packard Moreton. Minimum Curvature Variation Curves, Networks, and Surfaces for Fair Free-Form Shape Design . PhD thesis, University of California, Berkeley, Berkeley, California, 1992. Also issued as Technical Report UCB/CSD-93-732, March 1993

    work page 1992

  7. [7]

    Moreton and Carlo H

    Henry P. Moreton and Carlo H. S \'e quin. Functional optimization for fair surface design. In Proceedings of the 19th Annual Conference on Computer Graphics and Interactive Techniques , SIGGRAPH '92, pages 167--176, New York, NY, USA, July 1992. Association for Computing Machinery

    work page 1992

  8. [8]

    On the generalised ideal flow of closed planar curves

    James McCoy and Glen Wheeler. On the generalised ideal flow of closed planar curves. arXiv:2605.09379, May 2026

    work page Pith review arXiv 2026

Show all 17 references
  1. [9]

    Scale-critical curve diffusion flows

    Tatsuya Miura and Glen Wheeler. Scale-critical curve diffusion flows. arXiv:2604.01716, April 2026

    2026

  2. [10]

    A sixth order flow of plane curves with boundary conditions

    James McCoy, Glen Wheeler, and Yuhan Wu. A sixth order flow of plane curves with boundary conditions. Tohoku Mathematical Journal, Second Series , 72(3):379--393, September 2020

    2020

  3. [11]

    High order curvature flows of plane curves with generalised Neumann boundary conditions

    James McCoy, Glen Wheeler, and Yuhan Wu. High order curvature flows of plane curves with generalised Neumann boundary conditions. Advances in Calculus of Variations , 15(3):497--513, 2022. First published online 5 February 2021

    2022

  4. [12]

    McCoy, Glen E

    James A. McCoy, Glen E. Wheeler, and Yuhan Wu. A length-constrained ideal curve flow. The Quarterly Journal of Mathematics , 73(2):685--699, June 2022. First published online 15 November 2021

    2022

  5. [13]

    S. C. Ohlin. 2-D and 3-D Curve Interpolation by Consistent Splines . Internal report, IBM Nederland N.V., CAD/CAM Systems Support Group, Amsterdam, The Netherlands, March 1985

    1985

  6. [14]

    S. C. Ohlin. Splines for engineers. In Guy Mar \'e chal, editor, Eurographics '87: Proceedings of the European Computer Graphics Conference and Exhibition , pages 555--565, Amsterdam, The Netherlands, August 1987. Eurographics Association, North-Holland

    1987

  7. [15]

    The ideal flow for planar closed curves with local length constraint

    Shinya Okabe and Hikaru Yamaguchi. The ideal flow for planar closed curves with local length constraint. Advances in Nonlinear Analysis , 15(1):20250156, June 2026

    2026

  8. [16]

    Richard S. Palais. The principle of symmetric criticality. Communications in Mathematical Physics , 69(1):19--30, October 1979

    1979

  9. [17]

    Short time existence for higher order curvature flows with and without boundary conditions

    Yuhan Wu. Short time existence for higher order curvature flows with and without boundary conditions. In David R. Wood, Jan de Gier, Cheryl E. Praeger, and Terence Tao, editors, 2019--20 MATRIX Annals , volume 4 of MATRIX Book Series , pages 773--783. Springer, Cham, 2021

    2019

Pith tools

Reviewed August 27, 2026 · model on record in the stance chip above.