Pith. sign in

REVIEW 4 major objections 6 minor 38 references

On Courant-type bounds and spectral partitioning via Neumann domains on quantum graphs

T0 review · 4 major / 6 minor · reviewed 2026-08-04 · deepseek-v4-flash

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

Pith's one-line read On tree graphs, the nth eigenfunction has at most n−1 nodal points; under genericity, spectral minimal n-partition energies equal the (n+1)st eigenvalue.

desk verdict The tree node bound is probably true, but the paper's proof of it is broken in the middle; the partition theorems are plausible but also skip a key step. read the letter →

arxiv 2509.13228 v2 pith:HUP5MHRQ submitted 2025-09-16 math.SP math.AP

Luís Baptista , Matthias Hofmann This is my paper · ORCID
classification math.SPmath.AP MSC 34B4535P1535R0249Q1081Q35
keywords quantumgraphsLaplaciannodalcountNeumanndomainsspectralminimalpartitionstreeCourantboundgenericeigenfunctions
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 studies the Laplacian on quantum graphs that are trees and establishes two results. First, every eigenfunction belonging to the nth eigenvalue has at most n−1 nodal points, a Courant-type bound that holds without any genericity assumption. Second, when all eigenfunctions and spectral minimal partition minimizers are generic, the energy of the optimal spectral n-partition equals the (n+1)st Laplacian eigenvalue, L^N_n(G) = µ_{n+1}(G). Along the way the authors characterize when the Neumann domains of an eigenfunction form a spectral minimal partition. The examples show the results are sharp and that the genericity requirement is essential for the equality.

What carries the argument

The key machinery is a Courant counting argument adapted to metric graphs via 'cutting' at nodal points, combined with the surgery principle that allows cutting and gluing graphs without increasing certain eigenvalues. The notion of a generic (Morse) eigenfunction ensures each Neumann domain contains exactly one nodal point, making the Neumann-domain partition a candidate for the spectral minimal partition. The final equality L^N_n = µ_{n+1} follows from a gluing construction that reassembles the partition's eigenfunctions into a single eigenfunction with n nodal points.

What would settle it

Numerically search all tree graphs with up to five edges: if any eigenfunction of index n has at least n nodal points, the Courant-type bound fails. For the partition equality, compute L^N_n and µ_{n+1} on a tree where the nth eigenfunction vanishes at a vertex; a difference would show the genericity assumption is necessary.

Watch

Extended reading notes

Core claim

The paper claims two main theorems. Theorem 3.3: on a tree graph, an eigenfunction associated with the nth eigenvalue has at most n−1 nodal points. This is a Courant-type bound that does not require genericity; it is proved by a cutting argument that isolates the nodal component with the most nodal points and then uses domain monotonicity. Theorem 3.12: if all eigenfunctions of the tree are generic (simple, Morse, nonvanishing at vertices) and the spectral minimal partition admits generic minimizers, then L^N_n(G) = µ_{n+1}(G). The proof glues the restricted eigenfunctions of the partition pieces into a global eigenfunction, showing that the partition must be the Neumann partition of that ei

Load-bearing premise

The load-bearing premise is that all eigenfunctions and all spectral minimal partition minimizers are generic — simple eigenvalues, Morse functions, no vanishing at vertices — and, in the proof of the nodal bound, that certain restricted eigenfunctions on nested subgraphs are linearly independent.

Editorial extensions

If this is right

  • On tree graphs, the spectral minimal partition problem is solved by the Laplacian spectrum: L^N_n(G) = µ_{n+1}(G) whenever genericity holds.
  • The nodal count bound z_n ≤ n−1 holds for every tree eigenfunction, removing the usual genericity hypothesis from this Courant-type inequality.
  • Neumann domains of the nth eigenfunction form the optimal (n−1)-partition exactly when generic spectral minimizers exist, tying nodal geometry to spectral optimization.
  • If genericity fails, the equality can break, as Example 4.3 shows with a star graph and slightly unequal edge lengths, so the result marks the precise boundary of the phenomenon.

Reading between the lines

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

  • One could test whether the equality L^N_n = µ_{n+1} persists when only the minimizers are generic, not necessarily all eigenfunctions; the paper's examples leave this as an open possibility.
  • The nodal bound might extend to graphs with cycles by replacing n−1 with n−1+β, where β is the first Betti number; Example 4.2 shows the bound fails on a tadpole graph, suggesting such a correction.
  • Numerical sampling of random tree graphs could assess how often the genericity assumption actually holds and whether the spectral minimal partition energy is typically equal to the next eigenvalue.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 6 minor

Summary. The paper studies nodal and Neumann-domain structure of Laplacian eigenfunctions on compact metric quantum graphs, with emphasis on tree graphs. It claims three main results: (i) Theorem 3.3, an unconditional Courant-type bound asserting that on a tree an eigenfunction associated with μ_n has at most n−1 nodal points; (ii) Theorem 3.9, an equivalence between Neumann-domain partitions of the n-th eigenfunction and spectral minimal (n−1)-partitions under genericity assumptions; and (iii) Theorem 3.12, asserting that under full genericity of eigenfunctions and of spectral minimal minimizers, the Neumann spectral minimal n-partition energy satisfies L^N_n(G)=μ_{n+1}(G). The proofs combine a cutting/surgery framework, zero-extension arguments, and gluing of eigenfunctions on subgraphs. Several examples are given to illustrate sharpness and to show that genericity and the tree assumption are necessary.

Significance. If the main results are correct, the paper would establish a parameter-free nodal-point bound for all tree-graph eigenfunctions, removing the genericity assumptions that appear in earlier literature such as [Ber08, Sch06]. The spectral-partition identities under genericity would also give a concrete characterization of L^N_n(G) on generic trees, connecting Neumann-domain partitions to spectral minimal partitions. The paper contains useful examples, including a non-generic star-graph example showing that the partition equality can fail without genericity. However, the load-bearing proofs currently contain substantial gaps, so the significance can only be assessed after these are repaired.

major comments (4)
  1. [Theorem 3.3, Eqs. (3.7)-(3.10)] The contradiction proof is not completed. The subgraphs G^(j) are defined in (3.7) with a radius that does not depend on j, so no nesting or support ordering is specified. Consequently, the identity (3.9) for i≠j is not derived: the integration by parts is only valid with a precise statement of the boundary conditions at ∂G^(j) and of how the zero extensions are defined. The linear independence of ψ_1,...,ψ_m is asserted in one sentence; eigenfunctions of different Dirichlet problems on nested domains need not be linearly independent, and the argument does not rule out the possibility that all ψ_j are proportional to the same zero-extension of a ground state on G'. If independence fails, the m-dimensional subspace used in (3.10) is not available and the contradiction μ_m=μ_n is not obtained. This gap is load-bearing for the unconditional bound.
  2. [Theorem 3.12, gluing step] The proof glues the functions ψ_{2,i} on the clusters H_i. This is only possible if all μ_2(H_i) coincide; otherwise the glued function does not satisfy −u''=λu with a single λ. The definition of spectral minimal n-partition gives only max_i μ_2(H_i)=L^N_n(G), not equality of the individual values. The paper provides no proof that genericity forces the optimal partition to be an equipartition; indeed Example 4.3 shows that optimal partitions can fail to be equipartitions without genericity. Unless the equipartition property of the optimal partition is proved or added as an explicit hypothesis, the conclusion L^N_n(G)=μ_{n+1}(G) is unsupported.
  3. [Theorem 3.4] The proof is difficult to follow and contains a notational swap: it begins with 'some generic eigenfunction φ' and then 'Let ψ be an eigenfunction for μ_2(G)' before comparing the nodal point v of φ with the nodal point u of ψ. The final sentence invokes an undefined μ_j. More substantively, the domain-monotonicity comparison of Ĥ1 and Ĥ2 shows only a comparison of λ_1 values for two nodal domains; it does not establish that the originally given one-node Morse eigenfunction has eigenvalue μ_2. Since Theorem 3.7 uses Theorem 3.4, this proof needs to be rewritten carefully.
  4. [Lemma 3.2, Eqs. (3.2)-(3.5)] The min-max step in (3.4) is formally incorrect as written: it minimizes over 'f_1,...,f_n∈H^1_0(H)' rather than over n-dimensional subspaces, and it conflates λ_n(H) (Dirichlet on ∂H) with μ_n(G). Equation (3.3) also appears to have the scalar-product indices swapped. These issues are probably repairable, but because Lemma 3.2 feeds directly into Theorem 3.3, the repair should be explicit and included in a revised version.
minor comments (6)
  1. [Eq. (3.7)] The definition of G^(j) does not depend on j; if nested graphs with different radii are intended, the radii should be written as r_j, e.g. r_j = ℓ_min/(m+j).
  2. [Eq. (3.10)] The symbol m is used both as an arbitrary integer in the construction and as the index of μ_m in the min-max chain; this makes the direction of the final inequality hard to parse. Please clarify that the goal is to show μ_m ≤ μ_n for arbitrarily large m.
  3. [Example 4.2] The line 'L^N_k(G) ≤ L^N_k(I)=μ_k(I)' conflicts with Example 4.1, which gives L^N_k(I)=μ_{k+1}(I). The displayed equality should be corrected.
  4. [Theorem 3.12 proof] The proof refers to 'Theorem 3.5' and 'Theorem 3.1' where it appears to mean Proposition 3.5 and Lemma 3.1. Please correct the cross-references.
  5. [Example 4.2] The bullet list at the end repeats the same sentence twice: 'there may be no correspondence between the partitions formed by the Neumann domains of eigenfunctions and the spectral minimal partitions even if ...'. One occurrence should be removed.
  6. [Lemma 3.2, Eq. (3.3)] The displayed identity appears to have the indices swapped: it should likely be ⟨Σ_j α_j w_j', w_i'⟩ = μ_n ⟨Σ_j α_j w_j, w_i⟩. Please correct.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: Thm 3.3 has an unproved identity (3.9) — a correctness gap, not circularity; Thm 3.12 rests on explicit genericity assumptions and an external interlacing result. Self-citations are not load-bearing.

full rationale

The paper's derivation chain is essentially self-contained and does not reduce its conclusions to its inputs. The central upper bound (Theorem 3.3) uses Lemma 3.2, whose cross terms vanish because distinct nodal domains have disjoint interiors, and the min-max/domain-monotonicity chain (3.4)-(3.6) is standard. The subsequent identity (3.9), asserted for eigenfunctions on nested subgraphs G^(j), is not actually proved: zero-extending eigenfunctions from G^(j) to G can introduce boundary terms at ∂G^(j)\G', so (3.9) is a genuine proof gap. Likewise, the linear independence of ψ_1,...,ψ_m is merely asserted. However, an unsupported intermediate identity is a correctness concern, not circularity: it does not assume the target 'at most n−1 nodal points' conclusion. Theorem 3.12 also does not build its conclusion into its hypotheses: it explicitly assumes all eigenfunctions and the spectral minimal partition are generic, uses the published interlacing inequality (Proposition 3.11, from [HoKe21]) only for the lower bound, and constructs the (n+1)-th eigenfunction by gluing the cluster eigenfunctions; the equality L^N_n(G)=μ_{n+1}(G) is not an input. The self-citations to [HoKe21] and [HKMP21a] are peer-reviewed prior results used for standard surgery/interlacing facts and for examples; they are not invoked to forbid alternatives or to assume the target equalities. Examples 4.2 and 4.3 explicitly exhibit failure when tree/genericity assumptions break, confirming that the assumptions do real work rather than being renamed conclusions. No fitted parameter is renamed as a prediction, and no uniqueness theorem is imported from the authors. Overall circularity is therefore minimal.

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

The paper introduces no free parameters or new physical/mathematical entities. It relies on standard spectral theory of quantum graphs, Courant-type nodal bounds, and two domain assumptions: the tree structure and genericity. The genericity assumption is the most fragile; the examples show it is necessary for the partition equalities.

assumptions (5)
  • standard math Spectral theory of compact quantum graphs: discrete real spectrum, min-max characterization, surgery principles
    Invoked throughout Section 2 and in Lemma 2.9, Theorem 2.11, Lemma 3.2.
  • standard math Courant's nodal domain theorem for quantum graphs: the n-th eigenfunction has at most n nodal domains
    Not stated explicitly but underpins the node bound in Theorem 3.3; on a tree each node increases the number of nodal domains, which yields φ ≤ n-1.
  • domain assumption Genericity of all eigenfunctions and of spectral minimal partition minimizers (Definitions 2.5 and 3.8)
    Required in Theorems 3.9 and 3.12; Example 4.3 shows failure when eigenvalues have multiplicity and eigenfunctions are not generic.
  • domain assumption G is a tree graph (first Betti number zero)
    Used in Theorems 3.3, 3.7, 3.9, and 3.12; Example 4.2 shows the node bound fails when the graph contains a cycle.
  • ad hoc to paper Existence of generic spectral minimizers and an equipartition structure for the optimal partition
    Theorem 3.12 assumes this; Example 4.3 demonstrates that the optimal partition can fail to be an equipartition when eigenfunctions are non-generic.

how reviews work

0 comments
Cite this review

Pith. "Pith review of On Courant-type bounds and spectral partitioning via Neumann domains on quantum graphs." pith.science (2026). https://pith.science/paper/HUP5MHRQ

@misc{pith2026250913228,
  author       = {Pith},
  title        = {Pith review of: On Courant-type bounds and spectral partitioning via Neumann domains on quantum graphs},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/HUP5MHRQ}},
  note         = {Machine review of arXiv:2509.13228}
}
read the original abstract

We study the structure of eigenfunctions of the Laplacian on quantum graphs, with a particular focus on Morse eigenfunctions via nodal and Neumann domains. Building on Courant-type arguments, we establish upper bounds for the number of nodal points and explore conditions under which Neumann domains of eigenfunctions correspond to minimizers to a class of spectral partition problems often known as spectral minimal partitions. The main focus will be the analysis on tree graphs, where we characterize the spectral energies of such partitions and relate them to the eigenvalues of the Laplacian under genericity assumptions. Notably, we introduce a notion analogous to Courant-sharpness for Neumann counts and demonstrate when spectral minimal partitions coincide with partitions formed by Neumann domains of eigenfunctions.

Figures

Figures reproduced from arXiv: 2509.13228 by the authors.

Figure 3.1
Figure 3.1. A tree graph G (on the left) and the nodal points of ψ, in blue. On the left the cut of the graph trough v. Consider the graph that we obtain upon removal of Hn+1, then it contains n nodal domains (without loss of generality H1, . . . , Hn) such that each other nodal point is adjacent to two of these nodal domains. Denote G ′ a connected graph for which H1, . . . , Hn is an exhaustive partition, then by construction… view at source ↗
Figure 4.1
Figure 4.1. Path graph with nodal domains (left) and Neumann domains (right) of the eigenfunction to µ6 [PITH_FULL_IMAGE:figures/full_fig_p011_4_1.png] view at source ↗
Figure 4.2
Figure 4.2. Visualization of Graph G (left) for Example 4.2. The optimal Neumann partitions coincide with the ones in the interval with the same length (right) as con￾sidered in Example 4.1. Example 4.2. Consider the equilateral tadpole graph G shown in [PITH_FULL_IMAGE:figures/full_fig_p012_4_2.png] view at source ↗
Figures from the paper (6 more)
Figure 4.3
Figure 4.3. Figure 4.3: Plot of eigenfunctions to µI,1 on G via representation as a 1D domain for ℓ = 2π: loop from x = 0 to 2π, tail from x = 2π to 4π [PITH_FULL_IMAGE:figures/full_fig_p012_4_3.png]
Figure 4.4
Figure 4.4. Figure 4.4: Plot of the secular function and plots of the first three eigenfunctions via representation as a 1D domain for ℓ = 2π: loop from x = 0 to 2π, tail from x = 2π to 4π. The number of nodes of the eigenfunctions corresponding to the first five eigenfunctions are 0, 1, 3,…
Figure 4.5
Figure 4.5. Figure 4.5: Graph G of the example 4.3. Let us start by considering the equilateral case ϵ = 0. Then its eigenvalues are known (see [Fri05, Example 3]) to be (4.4) µ3j+1 = π 2 j 2 , µ3j+2 = µ3j+3 = π 2 (j + 1 2 ) 2 , j ∈ N0 [PITH_FULL_IMAGE:figures/full_fig_p014_4_5.png]
Figure 4.6
Figure 4.6. Figure 4.6: Visual representation of the nodal domains for the Morse eigenfunctions of µ7 = µ8 and µ9. White vertices denote the nodes of the eigenfunctions. These eigenvalues have multiplicites, and the corresponding eigenfunctions are not generic. In particular, some eigenfunc…
Figure 4.7
Figure 4.7. Figure 4.7: Neumann domains of the Morse eigenfunctions of µ7 = µ8 and µ9. The spectral minimal Neumann partitions for S3 are given by (see [HKMP21a, Lemma 7.4]) (4.5) L N 3j+1(S3) = π 2  j + 1 2 2 , L N 3j+2(S3) = L N 3j+3(S3) = π 2 (j + 1)2 [PITH_FULL_IMAGE:figures/full_fig…
Figure 4.8
Figure 4.8. Figure 4.8: Different setting of 2−partitions of the graph G. Let us describe the different configurations as shown in [PITH_FULL_IMAGE:figures/full_fig_p015_4_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

38 extracted references · 2 linked inside Pith

  1. [1]

    ``Neumann Domains on Graphs and Manifolds"

    Alon, L., Band, R., Bersudsky, M., Egger, S. ``Neumann Domains on Graphs and Manifolds". In: Keller M, Lenz D, Wojciechowski RK, eds. Analysis and Geometry on Graphs and Manifolds. London Mathematical Society Lecture Note Series. Cambridge University Press; 2020:203-249

    2020

  2. [2]

    L. Alon, R. Band and G. Berkolaiko, Nodal statistics on quantum graphs, Comm. Math. Phys. 362 (2018), no. 3, 909--948

    2018

  3. [3]

    Alon and R

    L. Alon and R. Band, Neumann domains on quantum graphs, Ann. Henri Poincar\'e 22 (2021), no. 10, 3391--3454

    2021

  4. [4]

    Arendt, W., Mazzeo, R., Friedlander's eigenvalue inequalities and the Dirichlet-to-Neumann semigroup, Comm.\ Pure Appl.\ Anal.\ 11 (2012), 2201--2212

    2012

  5. [5]

    Ariturk, S., Eigenvalue estimates on quantum graphs, preprint (2016), arXiv:1609.07471

    arXiv 2016

  6. [6]

    Band, R., L\'evy, G., Quantum graphs which optimize the spectral gap, Ann.\ Henri Poincar\'e 18 (2017), 3269--3323

    2017

  7. [7]

    Band, R., Berkolaiko, G., Raz, H., Smilansky, U., The number of nodal domains on quantum graphs as a stability index of graph partitions, Commun.\ Math.\ Phys.\ 311 (2012), 815--838

    2012

  8. [8]

    Band, R., Berkolaiko, G., Weyand, T., Anomalous nodal count and singularities in the dispersion relation of honeycomb graphs, J.\ Math.\ Phys.\ 56 (2015), 122111

    2015

Show all 38 references
  1. [9]

    Bogosel, B., Bonnaillie-Noël, V., Minimal partitions for p -norms of eigenvalues, Interfaces Free Bound.\ 20 (2018), 129--163

    2018

  2. [10]

    Berkolaiko, G., Kuchment, P., Introduction to Quantum Graphs, Mathematical Surveys and Monographs, vol.\ 186, American Mathematical Society, Providence, RI, 2013

    2013

  3. [11]

    Berkolaiko, G., A Lower Bound for Nodal Count on Discrete and Metric Graphs, Comm.\ Math.\ Phys.\ 278 (2008), 803--819

    2008

  4. [12]

    B., Kurasov, P., Mugnolo, D., Edge connectivity and the spectral gap of combinatorial and quantum graphs, J.\ Phys.\ A: Math.\ Theor.\ 50 (2017), 365201, 29pp

    Berkolaiko, G., Kennedy, J. B., Kurasov, P., Mugnolo, D., Edge connectivity and the spectral gap of combinatorial and quantum graphs, J.\ Phys.\ A: Math.\ Theor.\ 50 (2017), 365201, 29pp

    2017

  5. [13]

    B., Kurasov, P., Mugnolo, D., Surgery principles for the spectral analysis of quantum graphs, Trans.\ Amer.\ Math.\ Soc.\ 372 (2019), 5153--5197

    Berkolaiko, G., Kennedy, J. B., Kurasov, P., Mugnolo, D., Surgery principles for the spectral analysis of quantum graphs, Trans.\ Amer.\ Math.\ Soc.\ 372 (2019), 5153--5197

    2019

  6. [14]

    (ed.), Shape optimization and spectral theory, De Gruyter Open, Warsaw-Berlin, 2017

    Bonnaillie-Noël, V., Helffer, B., Nodal and spectral minimal partitions -- The state of the art in 2016, Chapter 10 in Henrot, A. (ed.), Shape optimization and spectral theory, De Gruyter Open, Warsaw-Berlin, 2017

    2016

  7. [15]

    Bonnaillie-Noël, V., L\'ena, C., Spectral minimal partitions for a family of tori, Exp.\ Math.\ 26 (2017), 381--395

    2017

  8. [16]

    Courant, R., Ein Allgemeiner Satz Zur Theorie Der Eigenfunktionen Selbsadjungierter Differentialausdrücke. Nachr. Ges. Wiss. Göttingen, Math.-Phys. Kl. 1923(1) (1923), 81–84

    1923

  9. [17]

    Conti, M., Terracini, S., Verzini, G., On a class of optimal partition problems related to the Fu c ík spectrum and to the monotonicity formulae , Calc.\ Var.\ Partial Differential Equations 22 (2005), 45--72

    2005

  10. [18]

    Filonov, N., On an inequality for the eigenvalues of the Dirichlet and Neumann problems for the Laplace operator, St.\ Petersburg Math.\ J.\ 16 (2005), 413--416

    2005

  11. [19]

    Friedlander, L., Some Inequalities between Dirichlet and Neumann Eigenvalues, Arch.\ Rat.\ Mech.\ Anal.\ 116 (1991), 153--160

    1991

  12. [20]

    Friedlander, L., Extremal properties of eigenvalues for a metric graph, Ann.\ Inst.\ Fourier 55 (2005), 199--212

    2005

  13. [21]

    Helffer, B., Hoffmann-Ostenhof, T., Terracini, S., Nodal domains and spectral minimal partitions, Ann.\ Inst.\ Henri Poincar\'e (C) Anal.\ Non Lineaire 26 (2009), 101--138

    2009

  14. [22]

    B., Mugnolo, D., Plümer, M., Asymptotics and Estimates for Spectral Minimal Partitions of Metric Graphs, Int.\ Equ.\ Oper.\ Theory 93 (2021), 26

    Hofmann, M., Kennedy, J. B., Mugnolo, D., Plümer, M., Asymptotics and Estimates for Spectral Minimal Partitions of Metric Graphs, Int.\ Equ.\ Oper.\ Theory 93 (2021), 26

    2021

  15. [23]

    B., Mugnolo, D., Plümer, M., On Pleijel’s Nodal Domain Theorem for Quantum Graphs

    Hofmann, M., Kennedy, J. B., Mugnolo, D., Plümer, M., On Pleijel’s Nodal Domain Theorem for Quantum Graphs. Ann. Henri Poincaré (2021), 1–30

    2021

  16. [24]

    B., Interlacing and Friedlander-type inequalities for spectral minimal partitions of metric graphs, Lett

    Hofmann, M., Kennedy, J. B., Interlacing and Friedlander-type inequalities for spectral minimal partitions of metric graphs, Lett. Math. Phys. 111 (2021), no. 4, Paper No. 96, 30 pp

    2021

  17. [25]

    B., Geometric spectral theory of quantum graphs, Commun.\ Math.\ 32 (2024), no

    Kennedy, J. B., Geometric spectral theory of quantum graphs, Commun.\ Math.\ 32 (2024), no. 3, Special issue: Portuguese Mathematics, Paper no. 12380

    2024

  18. [26]

    B., Kurasov, P., L\'ena, C., Mugnolo, D., A theory of spectral partitions of metric graphs, Calc.\ Var.\ Partial Differential Equations 60 (2021), 61

    Kennedy, J. B., Kurasov, P., L\'ena, C., Mugnolo, D., A theory of spectral partitions of metric graphs, Calc.\ Var.\ Partial Differential Equations 60 (2021), 61

    2021

  19. [27]

    B., Kurasov, P., Malenov\'a, G., Mugnolo, D., On the spectral gap of a quantum graph, Ann.\ Henri Poincar\'e 17 (2016), 2439--2473

    Kennedy, J. B., Kurasov, P., Malenov\'a, G., Mugnolo, D., On the spectral gap of a quantum graph, Ann.\ Henri Poincar\'e 17 (2016), 2439--2473

    2016

  20. [28]

    and Smilansky, U., Periodic orbit theory and spectral statistics for quantum graphs, Ann

    Kottos, T. and Smilansky, U., Periodic orbit theory and spectral statistics for quantum graphs, Ann. Physics 274 (1999), no. 1, 76--124

    1999

  21. [29]

    Rep.\ Math.\ Phys.\ 82 (2018), 63--80

    Kurasov, P., Serio, A., On the sharpness of spectral estimates for graph Laplacians. Rep.\ Math.\ Phys.\ 82 (2018), 63--80

    2018

  22. [30]

    Lett Math Phys 109, 2491–2512 (2019)

    Kurasov, P., On the ground state for quantum graphs. Lett Math Phys 109, 2491–2512 (2019)

    2019

  23. [31]

    Kurasov, P., Spectral geometry of graphs , Operator Theory: Advances and Applications, 293, Birkh\"auser/Springer, Berlin, 2024

    2024

  24. [32]

    Mugnolo, D., What is actually a metric graph?, preprint (2019), arXiv:1912.07549

    2019 arXiv

  25. [33]

    Nicaise, S., Spectre des r\'eseaux topologiques finis, Bull.\ Sci.\ Math.\ (2) 111 (1987), 401--413

    1987

  26. [34]

    Comm.\ Pure Appl.\ Math., 9 (1956), 543–550

    Pleijel, Å., Remarks on Courant’s nodal line theorem. Comm.\ Pure Appl.\ Math., 9 (1956), 543–550

    1956

  27. [35]

    Rohleder, J., Eigenvalue Estimates for the Laplacian on a Metric Tree, Proc.\ Amer.\ Math.\ Soc.\ 145 (2017), 2119--2129

    2017

  28. [36]

    3, 167--178

    Schapotschnikow, P., Eigenvalue and nodal properties on quantum graph trees, Waves Random Complex Media 16 (2006), no. 3, 167--178

    2006

  29. [37]

    J.\ Phys.\ A: Math.\ Theor.\ 54 (2021), 015202, 14 pp

    Serio, A., On extremal eigenvalues of the graph Laplacian. J.\ Phys.\ A: Math.\ Theor.\ 54 (2021), 015202, 14 pp

    2021

  30. [38]

    Journal de Mathématiques Pures et Appliquées (1836), 106–186

    Sturm, C., Mémoire sur les équations différentielles linéaires du second ordre. Journal de Mathématiques Pures et Appliquées (1836), 106–186

Pith tools

Reviewed August 4, 2026 · model on record in the stance chip above.