Invariant Hypersurfaces and Nodal Components of Foliations

Felipe Cano; Jean Fran\c{c}ois Mattei; Marianna Ravara-Vago

REVIEW 2 major objections 5 minor 26 references

Invariant Hypersurfaces and Nodal Components of Foliations

T0 review · 2 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read After resolving a singular foliation, every region cut out by nodal barriers contains an invariant hypersurface.

desk verdict New conditional result: the nodal separating block counting and invariant hypersurface statement are genuinely new, but the 'any dimension' claim rests on an unproved desingularization assumption that the authors explicitly flag. read the letter →

arxiv 1908.08324 v1 pith:KZJ7ERY5 submitted 2019-08-22 math.DS math.AG

Felipe Cano , Jean François Mattei , Marianna Ravara-Vago This is my paper · ORCID

classification math.DSmath.AG MSC 37F7532S6514B0532S45

keywords singularholomorphicfoliationsinvarianthypersurfacesnodalseparatingblocksreductionofsingularitiesgeneralizedhypersurfaceseparatricesexceptionaldivisorcombinatoricsinhigherdimension

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 a higher-dimensional analogue of the two-dimensional result that, after a nodal reduction of singularities of a holomorphic foliation germ, every connected component of the exceptional divisor minus the nodal points meets an invariant curve. For a non-dicritical complex hyperbolic (GH) foliation on $(\mathbb{C}^n,0)$, the paper shows that each connected component $C$ of $E \setminus |S|$ is met by the strict transform of a genuine invariant hypersurface of the original germ, and that this strict transform meets the divisor only inside $C$. A second result counts the regions: if the nodal separator set has $s$ nodal separating blocks, then $E \setminus |S|$ has exactly $s+1$ connected components. The interest is that the leaf-space separation observed in dimension two persists in every ambient dimension, with invariant hypersurfaces witnessing each region.

What carries the argument

The load-bearing objects are nodal separating blocks and combinatorial strata structures. A nodal separating block is a connected union of irreducible components of the singular locus of generic nodal type that meets only nodal and real saddle components; after a nodal reduction these blocks sit at corners of the exceptional divisor and act as barriers in the space of leaves. The separator set $S$ is their union, and the complement $E\setminus|S|$ is studied through the combinatorial strata structure $H_M$ of the divisor, an abstract simplicial complex whose $1$-connected components are the connected components of the divisor. The counting mechanism is Theorem 3: on a simply connected strata structure with a datum of nodal strata, the number of $1$-connected components of the structure minus the separator closure is $n+1$, proved by a parity argument counting how many times a path crosses a given separating block, together with a combinatorial Seifert–van Kampen theorem. The ambient dimension is reduced by two-equireduction points, where a chosen transverse surface inherits a genuine two-dimensional reduction of singularities, so the known two-dimensional refined theorem can be applied at the end.

What would settle it

A concrete counterexample would be a GH-foliation on $(\mathbb{C}^4,0)$ with a nodal reduction for which some connected component $C$ of $E\setminus|S|$ fails to meet any invariant hypersurface, contradicting Theorem 1; alternatively, exhibiting a four-dimensional pair $(M_0,L_0)$ of an ambient space and a finite list of irreducible hypersurfaces that admits no admissible reduction of singularities would remove the stated domain of the theorems, since the existence of nodal reductions is derived from that assumption.

Watch

Extended reading notes

Core claim

The central claim is Theorem 1: for any nodal reduction $\pi:(M,E,F)\to((\mathbb{C}^n,0),\varnothing,F_0)$ of a GH-foliation, if $|S|$ is the support of the nodal separator set, every connected component $C$ of $E\setminus|S|$ has an invariant hypersurface $H_0$ of $F_0$ whose strict transform $H$ satisfies $H\cap C\neq\varnothing$ and $H\cap E\subset C$. The paper derives this from the more general Theorem 5, which allows an initial divisor $E_0$ and states that each component $C$ either contains the strict transform of an irreducible component of $E_0$ or contains a trace-type point through which an invariant branch passes. The proof joins the two-dimensional refined theorem on invariant branches, a two-equireduction argument that reduces the ambient dimension, and a purely combinatorial counting result (Theorem 3) showing that each nodal separating block splits one connected component of the divisor complement into two. The same count gives Theorem 2: $s$ separating blocks produce exactly $s+1$ components of $E\setminus|S|$.

Load-bearing premise

The argument assumes that every finite collection of hypersurfaces in any dimension can be resolved by a sequence of admissible blow-ups; the paper outlines this only in dimension three and notes that dimension four and higher have no explicit statement in the literature.

Editorial extensions

If this is right

For a GH-foliation germ, the number of invariant hypersurfaces that meet the exceptional divisor is at least the number $s+1$ of regions cut out by the nodal separator set, so no region between nodal barriers can be empty.
Because the strict transform of each found hypersurface meets the divisor only inside its own region, invariant hypersurfaces never cross a nodal separating block, matching the leaf-space separation seen at nodal points.
Theorem 2 gives the exact count: if exactly $s$ nodal separating blocks appear, the divisor complement has exactly $s+1$ connected components, so in any dimension the separator set behaves like $s$ nodal points in the plane.
Through the bijection between invariant hypersurfaces of the desingularized space and those of $F_0$, each region yields a genuine irreducible invariant germ $(H_0,0)\subset(\mathbb{C}^n,0)$, not merely a local or formal hypersurface.
The two-equireduction argument shows the higher-dimensional statement is ultimately a consequence of the two-dimensional refined theorem applied along carefully chosen transverse surfaces.

Reading between the lines

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

If the assumed admissible-reduction statement fails in dimension $\geq 4$, the theorem's stated domain shrinks, but the conclusion could still hold for every concrete nodal reduction that exists, since the proof uses the reduction only through its combinatorial strata structure and two-equireduction properties.
The parity counting behind Theorem 3 suggests viewing the separator set as a cut locus in the dual complex of the divisor; one could ask whether a Morse-theoretic or Euler-characteristic proof of the $s+1$ count exists without the simple-connectivity hypothesis.
At a nodal separating block, local linearization supplies a real invariant hypersurface separating leaves; a testable extension would be to use these local real barriers to name the complex invariant hypersurface assigned to each component $C$, instead of proving existence only.
The same combinatorial counting may extend to dicritical or saddle-node foliations if 'nodal separating blocks' are replaced by a broader class of separating strata, although the paper explicitly restricts to GH-foliations.

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.

Desk Editor's Note

read the letter

Read the full text. Theorems 1 and 2 are new relative to the cited literature—[24] gives the two-dimensional base case, and the component-wise statement plus the s+1 count are not in any of the references. The proof has two separate halves. The combinatorial-topological half (Sections 6–7) is the strongest part: the parity lemma, the counting via the Φ_m maps, and the blow-up stability of simple connectedness are written out in detail and internally consistent. The geometric half (Section 8) is more compressed; it reduces to dimension two by two-equireduction and leans on the authors' own previous work [4]. That is acceptable if the cited base case holds, but it means a referee needs to trust those steps.

The soft spot is the one the paper itself points at. Remark 2 assumes, without proof, that admissible reductions of singularities exist for hypersurface pairs in arbitrary dimension, and Proposition 2 turns that into existence of nodal reductions. Theorems 1 and 2 are then stated for those reductions. The authors give an outline for dimension three and say explicitly that for n ≥ 4 the literature has no such statement. So the abstract's "any ambient dimension" is an overstatement: what is established is conditional on a strong desingularization result. This is not an internal contradiction, and it is not a manufactured flaw; it is just where the paper stands. If the desingularization result is supplied, the theorems carry through.

I do not see a circularity problem. The base case is external [24], and the higher-dimensional proof reduces to it by restriction to transversals. The counting results are independent of the geometric input. The reliance on prior work is heavy, but it is concentrated in background that is cited, not reinvented.

Who gets value: people working on desingularization of foliations and separatrix existence in higher dimension. This is a serious paper with a real new result, conditional on an open input it openly names. It deserves a serious referee, but the referee's first job is to verify whether the Remark 2 assumption is truly missing or can be imported from Hironaka-style resolution with the connectivity condition. If the authors can prove it, or restate the theorems conditionally in the abstract, I would take the theorems.

Referee Report

2 major / 5 minor

Summary. The manuscript generalizes to arbitrary ambient dimension the two-dimensional theorem of Ortiz-Rosales-Voronin: after a nodal reduction of singularities of a non-dicritical complex-hyperbolic (GH) foliation on (C^n,0), each connected component of the exceptional divisor minus the nodal separator set is met by the strict transform of an invariant hypersurface whose intersection with the divisor lies in that component (Theorem 1). It also proves that if the separator set has s nodal separating blocks, the complement has exactly s+1 connected components (Theorem 2). The main tools are: nodal separating blocks and nodally reduced GH-foliated spaces (Section 5); a purely combinatorial theorem counting connected components for simply connected strata structures with a datum of nodal strata (Theorems 3, Sections 6-7); and a reduction to the two-dimensional case via two-equireduction (Section 8). The paper is explicit that the existence of nodal reductions is taken as an assumption (Remark 2, Section 3), with only an outline of the required desingularization in dimension three (Appendix 10).

Significance. If the standing assumptions are met, the results provide a natural higher-dimensional counterpart of the refined Camacho-Sad theorem and a clean statement that nodal separating blocks separate the exceptional divisor exactly like nodal points in dimension two. The combinatorial core is worked out in detail: the parity lemma (Lemma 4), the component-counting Proposition 6, and the stability of simple connectedness under blow-up (Section 7.3) are plausible and internally consistent. The paper is also honest about its main standing assumption, which is a genuine strength of the exposition. The value of the paper would be substantially increased if the admissible-desingularization premise were proved, or if the statements were made explicitly conditional on it.

major comments (2)

[§3, Remark 2; §5.1; §10] Theorems 1 and 2 are stated unconditionally in the abstract and introduction, but their domain of validity is exactly the domain of the unproved admissible-desingularization statement in Remark 2. Proposition 2 derives the existence of reductions of singularities of GH-foliations from that statement, and Proposition 3 converts any such reduction into a nodal reduction; hence the very objects to which Theorems 1 and 2 apply may fail to exist in dimension n >= 4 if the assumption fails. The paper itself notes that the statement is only outlined in dimension three (Appendix 10) and that 'there are no explicit statements in the literature' for n >= 4. This is load-bearing rather than cosmetic. Please either prove the admissible-desingularization statement, cite a published theorem that implies it, or restate the main results explicitly as conditional on this assumption and adjust the abstract accordingly.
[§8.2, Propositions 9 and 10] The reduction to dimension two, which is the core of the proof of Theorem 1, depends on Propositions 9 and 10, but their proofs are not given in the manuscript: the text says the proof is 'essentially contained in [4]' and 'See [4]'. Since the present paper claims a result in arbitrary ambient dimension and uses the framework of standard ambient spaces and nodal separator sets, the precise statements in [4] that imply Propositions 9 and 10 should be identified, or the proofs should be included. Without this, the induction in Section 8.4 is not self-contained and the applicability of the two-dimensional theorem [24, Corollary 4.1] cannot be verified from the material presented.

minor comments (5)

[§5.2, Example 2] The text contains an unresolved cross-reference 'see Remark ??'; this should be replaced by the number of the intended remark.
[§8.1, Remark 16] The notation 'Ei ⊂ C' for a connected component C of the open set E \ |S| is inconsistent; since C is open in E, the correct formulation should be 'Ei ∩ C ≠ ∅' or 'Ei ⊂ \overline{C}'.
[§1, Theorem 2] In the statement of Theorem 2, 'separator set of (M, E, ~F)' contains a stray tilde before F; it should read '(M, E, F)'.
[§5.1, Proposition 3] There is a typo in the proof: 'admisible transformation' should be 'admissible transformation'.
[§7.2] The subsection title 'Combinatorial Strata Sructure After Blow-up' contains a typo; 'Sructure' should be 'Structure'.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the invariant-hypersurface theorem is an induction whose base case is the external 2D result [24]; the higher-dimensional setting rests on an openly stated desingularization assumption (Remark 2) that is a domain caveat, not a circular derivation.

full rationale

Theorems 1 and 2 are not obtained from their own conclusions. Theorem 2 is proved by a self-contained combinatorial-topological counting argument (Sections 6–7) based on the divisor stratification; Theorem 1 is proved by induction and by reduction to the external two-dimensional theorem [24, Cor. 4.1] through strict transversals (Sections 8.2–8.4). The authors cite their own prior work [2–6] for definitions, simple points, partial separatrices, and equireduction facts, but these are background results and are not used to assume the target statement. The only substantive caveat is Remark 2: nodal reductions in all dimensions exist only under an explicitly unproved admissible-desingularization assumption, with only an outline in dimension three and no statement for n ≥ 4. This is load-bearing for the domain of application of the theorems, but it is an input hypothesis, not a disguised version of the conclusion, so it is not circularity. The base case [24] is external, and the proof structure gives independent content.

Assumptions & free parameters 0 free parameters · 7 assumptions · 2 invented entities

This is a pure mathematics paper: there are no fitted parameters, no data, and no empirical inputs. The central claim rests on seven axioms: the explicitly assumed admissible resolution of singularities in arbitrary dimension (Remark 2), the external 2D base theorem [24], the equireduction results quoted from the authors' own [4], the desingularization and separatrices theory from [2,3,13], local linearization of nodal singularities [10,11], and standard algebraic topology facts. The first is the fragile one: it is stated for every dimension but explicitly unproved for n >= 4, and only sketched for n = 3. The invented entities (nodal separating blocks, datum of nodal strata) are definitions, not empirical postulates.

assumptions (7)

ad hoc to paper Admissible reduction of singularities exists for every standard ambient pair (M0, L0) in any dimension (Remark 2, Section 3).
The paper explicitly works 'under the assumption that the following statement is true'. Used via Proposition 2 to obtain reductions of singularities of GH-foliations; the nodal reductions to which Theorems 1 and 2 apply exist only if it holds. Outlined for dimension 3 in Appendix 10; no statement in the literature for n >= 4.
domain assumption Two-dimensional refined Camacho-Sad theorem of Ortiz-Rosales-Voronin [24, Corollary 4.1].
Invoked in Section 8.4 as the base case: the restriction of the reduction to a suitable two-dimensional transversal inherits a 2D nodal reduction, and [24] guarantees an invariant branch through each component of the complement of nodal points.
domain assumption Two-equireduction results, Propositions 9 and 10, 'essentially contained in [4]' (Cano-Mattei).
Section 8.2 states that the adapted singular locus minus the equireduction points has codimension >= 3 (Proposition 9) and that transversals to equireduction points yield genuine 2D reductions (Proposition 10). These are quoted, not proved.
domain assumption The desingularization and partial separatrix theory of non-dicritical codimension one foliations from [2], [3], [4] and [13] (Propositions 1 and 2, Remark 5, Section 4).
The characterization of simple points (Prop. 1), the fact that resolution of invariant hypersurfaces resolves the foliation (Prop. 2), and the bijection between partial separatrices and invariant hypersurfaces (Remark 5, properties in [3]) are taken from these works.
domain assumption Local linearization of nodal-type singularities (Remark 3, citing [10] and [11]).
Used in Section 4.1 and Lemma 1 to write local normal forms of nodal points, to identify the coordinate hyperplanes that belong to the divisor E, and to classify the components xi = xj = 0 as nodal or real saddle type.
standard math Standard algebraic topology facts: Seifert-van Kampen, connectedness of simplicial complexes, fundamental group of a complex with a 2-simplex removed (Sections 7 and 9, citing [18] and [21]).
Used in the proof of Theorem 4 (stability of simple connectedness under blow-up) and Proposition 8 (equivalence between combinatorial and topological simple connectedness).
standard math Grauert's Direct Image Theorem (Remark 5).
Used to transfer the bijection between invariant hypersurfaces of the reduced space and invariant hypersurfaces of the original germ after reduction.

invented entities (2)

Nodal separating blocks and nodal separator set S (Definition 8, Section 5)
purpose: Define the higher-dimensional analogue of 2D nodal points; Theorems 1 and 2 are statements about the complement E minus |S| and about the number of blocks of S.
New mathematical objects introduced in this paper (building on 'uninterrupted nodal components' in [5,6]). They are characterized internally (Lemmas 1-2, Proposition 4) but carry no external falsifiable prediction.
Datum of nodal strata (N, {P_J}) on a combinatorial strata structure (Section 6.2)
purpose: Abstract encoding of which divisor strata are nodal and which of their two local branches have the same sign; used in the proof of the s+1 counting theorem (Theorem 3).
Auxiliary combinatorial formalism introduced for the proof; no empirical content, verified only through the internal lemmas of Sections 6-7.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Invariant Hypersurfaces and Nodal Components of Foliations." pith.science (2026). https://pith.science/paper/KZJ7ERY5

@misc{pith2026190808324,
  author       = {Pith},
  title        = {Pith review of: Invariant Hypersurfaces and Nodal Components of Foliations},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KZJ7ERY5}},
  note         = {Machine review of arXiv:1908.08324}
}

abstract

It is known that there is at least an invariant analytic curve passing through each of the components in the complement of nodal singularities, after the reduction of singularities of a germ of singular foliation in ${\mathbb C}^2,0$}. Here, we state and prove a generalization of this property to any ambient dimension.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

26 extracted references · 25 canonical work pages

[24]

Moscow Mathematical Journal, V 12 (2012), no.4, 825-862

Ortiz-Bobadilla, L.; Rosales-Gonz´ alez, E.; Voronin , S.: Thom’s problem for generic degener- ated singular points of holomorphic foliations in the plane . Moscow Mathematical Journal, V 12 (2012), no.4, 825-862

work page 2012
[4]

Annales de l’Institut Fourier, 42, 1-2, (1992), 49-72

Cano, F.; Mattei, J.F.: Hypersurfaces int´ egrales de feuilletages holomorphes. Annales de l’Institut Fourier, 42, 1-2, (1992), 49-72

work page 1992
[1]

Ergebnisse der Mathematik und Ihrer Grenzgebiete

Aroca, J.M., Hironaka, H., Vicente, J.L.: Complex Analytic Desingularization. Ergebnisse der Mathematik und Ihrer Grenzgebiete. 3 Folge /A Series of Mode rn Surveys in Mathematics. Springer Verlag, Japan, 2018. 330 p.. ISBN-10: 4431702180

work page 2018
[2]

Annals of Mathematics

Cano, F.: Reduction of the singularities of codimension one singular foliations in dimension three. Annals of Mathematics. 160, (2004), 907-1011

work page 2004
[3]

Acta Mathematica, 169, (1992), pp

Cano, F.; Cerveau, D.: Desingularization of non-dicritical holomorphic foliati ons and exis- tence of separatrices. Acta Mathematica, 169, (1992), pp. 1-103

work page 1992
[5]

Partial Separatrices

Cano, F.; Ravara-Vago, M.: Local Brunella’s Alternative II. Partial Separatrices. Interna- tional Mathematics Research Notices, 40 pages. (2015)

work page 2015
[6]

RICH Foliations

Cano, F.; Ravara-Vago, M.; Soares, M.: Local Brunella’s Alternative I. RICH Foliations. . International Mathematics Research Notices, 50 pages. (20 15) Vol. 2015, 2525-2575

work page 2015
[7]

Camacho, C.; Lins-Neto, A.; Sad, P.: Topological invariants and equidesingularization for holomorphic vector ﬁelds. J. Diﬀerential Geom., 20, 1 (1984), 143–174

work page 1984

Show all 26 references

[8]

Camacho, C.; Sad, P.: Invariant varieties through singularities of holomorphic vector ﬁelds. Ann. of Math. (2), 115, 3 (1982), 579–595

1982
[9]

Ergodic Theory and Dynamical Systems, 8, 36, (2015), 2408-2418

Camacho, C.; Rosas, R.: Invariant sets near singularities of holomorphic foliatio ns. Ergodic Theory and Dynamical Systems, 8, 36, (2015), 2408-2418

2015
[10]

Cerveau,D.; Lins-Neto, A: Formes tangentes ` a des actions commutatives. Ann. Fac. Sci. Toulouse Math. (5) 6 (1984), no. 1, 51–85

1984
[11]

Ast´ erisque, 97

Cerveau,D.; Mattei, J.F.: Formes int´ egrables holomorphes singuli` eres. Ast´ erisque, 97. Soci´ et´ e Math´ ematique de France, Paris, 1982. 193 pp

1982
[12]

Springer

Cossart, V.; Giraud, J.; Orbanz, U.: Resolution of Surface Singularities (Three lectures and an Appendix with Hironaka’s Bowdoin college Seminar)Lecture Notes in Mathematics. Springer. ISBN 978-3-540-39125-8. 1984

1984
[13]

Ast´ erisque 319, (2008), 261-268

Fern´ andez, P.; Mozo, J.: On generalized surfaces in (C3,0). . Ast´ erisque 319, (2008), 261-268

2008
[14]

Wiley-Interscience, 1994

Griﬃths, P.; Harris, J.: Principles of Algebraic Geome try. Wiley-Interscience, 1994

1994
[15]

Hironaka, H.: Resolution of singularities of an algebraic variety over a ﬁ eld of characteristic zero. Ann. of Math. 79 (1964), 109–326

1964
[16]

Hironaka, H.: Certain Numerical Characters of Singularities. J. Math. Kyoto Univ. 10 (1970), no. 1, 151–187. doi:10.1215/kjm/125 0523817. https://projecteuclid.org/euclid.kjm/1250523817

1970
[17]

Ed Springer-Verlag, Lecture Notes in Math- ematics 708 (1979)

Jouanolou, J.P.: Equations de Pfaﬀ Alg´ ebriques. Ed Springer-Verlag, Lecture Notes in Math- ematics 708 (1979)

1979
[18]

Springer

Kozlov, D.: Combinatorial Algebraic TopologyAlgorithms and Computations in Mathematics, vol 21. Springer. 2008 INV ARIANT HYPERSURF ACES AND NODAL COMPONENTS OF FOLIATION S 33

2008
[19]

Ann.Sci.Ec

Mattei, J.F.; Mar ´ ın, D.:Incompressibilit´ e des feuilles de germes de feuilletages holomorphes singuliers. Ann.Sci.Ec. Norm. Sup. 41, no. 6 (2008): 855–903

2008
[20]

Journal of Singularities volume 9 (2014), 122-150

Mattei, J.F.; Mar ´ ın, D.:Topology of singular holomorphic foliations along a compac t divisor. Journal of Singularities volume 9 (2014), 122-150

2014
[21]

New York

Massey, W.S.: Algebraic Topology: an introduction Harcourt Brace and W orld Inc. New York. 1967

1967
[22]

4 (1980): 469–523

Mattei, J.F.; Moussu, R.: Holonomie et int´ egrales premi` eres.Annales Scientiﬁques de l’ ´Ecole Normale Sup´ erieure 13, no. 4 (1980): 469–523

1980
[23]

Revista de la Real Academia de Ciencias Exactas, F ´ ısicas y Naturales

Molina, B.: Combinatorial Aspects of Classical Resolution of Singular ities. Revista de la Real Academia de Ciencias Exactas, F ´ ısicas y Naturales. Se rie A, Matem´ aticas (2017). To appear. 10.1007/s13398-018-0583-z. Molina-Samper, Beatriz. (2017). Combinatorial Aspects of ...

2017 doi
[25]

American Journal of Mathematics 90, no

Seidenberg, A.: Reduction of singularities of the diﬀerential equation Ady = Bdx. American Journal of Mathematics 90, no. 1 (1968) 248–69

1968
[26]

Villamayor, O.: Constructiveness of Hironaka’s Resolution of Singulariti es. Ann. Sci. ´Ecole Norm. Sup. 22 (1989), 1–32

1989

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [24]

Moscow Mathematical Journal, V 12 (2012), no.4, 825-862

Ortiz-Bobadilla, L.; Rosales-Gonz´ alez, E.; Voronin , S.: Thom’s problem for generic degener- ated singular points of holomorphic foliations in the plane . Moscow Mathematical Journal, V 12 (2012), no.4, 825-862

work page 2012

[2] [4]

Annales de l’Institut Fourier, 42, 1-2, (1992), 49-72

Cano, F.; Mattei, J.F.: Hypersurfaces int´ egrales de feuilletages holomorphes. Annales de l’Institut Fourier, 42, 1-2, (1992), 49-72

work page 1992

[3] [1]

Ergebnisse der Mathematik und Ihrer Grenzgebiete

Aroca, J.M., Hironaka, H., Vicente, J.L.: Complex Analytic Desingularization. Ergebnisse der Mathematik und Ihrer Grenzgebiete. 3 Folge /A Series of Mode rn Surveys in Mathematics. Springer Verlag, Japan, 2018. 330 p.. ISBN-10: 4431702180

work page 2018

[4] [2]

Annals of Mathematics

Cano, F.: Reduction of the singularities of codimension one singular foliations in dimension three. Annals of Mathematics. 160, (2004), 907-1011

work page 2004

[5] [3]

Acta Mathematica, 169, (1992), pp

Cano, F.; Cerveau, D.: Desingularization of non-dicritical holomorphic foliati ons and exis- tence of separatrices. Acta Mathematica, 169, (1992), pp. 1-103

work page 1992

[6] [5]

Partial Separatrices

Cano, F.; Ravara-Vago, M.: Local Brunella’s Alternative II. Partial Separatrices. Interna- tional Mathematics Research Notices, 40 pages. (2015)

work page 2015

[7] [6]

RICH Foliations

Cano, F.; Ravara-Vago, M.; Soares, M.: Local Brunella’s Alternative I. RICH Foliations. . International Mathematics Research Notices, 50 pages. (20 15) Vol. 2015, 2525-2575

work page 2015

[8] [7]

Camacho, C.; Lins-Neto, A.; Sad, P.: Topological invariants and equidesingularization for holomorphic vector ﬁelds. J. Diﬀerential Geom., 20, 1 (1984), 143–174

work page 1984