REVIEW 3 major objections 1 minor 11 references
A Fibonacci theorem for Collatz trajectories via modular graph structure
T0 review · 3 major / 1 minor · reviewed 2026-06-29 · grok-4.3
Pith's one-line read Exactly F(m+1) odd integers up to 2^m avoid 4 mod 6 in their Collatz steps 2 to m.
desk verdict The main claim fails a direct check for m=3, where only 2 odds up to 8 avoid 4 mod 6 instead of the stated 3. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The directed graph G of Collatz transitions modulo 6 and the spectral radius ϕ of the subgraph obtained by removing vertex 4 from its absorbing strongly connected component G'.
What would settle it
For m=3, enumerate the four odd integers up to 8 and count how many avoid residue 4 mod 6 during steps 2 and 3; the result must equal F(4)=3 if the claim holds.
Extended reading notes
Core claim
We prove that for each m≥1, exactly F(m+1) odd integers in {1,…,2^m} have the property that their orbit under T avoids the residue class 4 mod6 during steps 2,…,m, where F(m+1) is the (m+1)-th Fibonacci number; the proportion decays at rate (ϕ/2)^m, ϕ=(1+√5)/2. The proof uses the directed graph G of Collatz transitions modulo 6 and its unique absorbing strongly connected component G'={1,2,4,5}. Removing vertex 4 from G' yields a subgraph of spectral radius ϕ, against ρ(G')=2; the Fibonacci count follows from this spectral gap.
Load-bearing premise
The absorbing component of the Collatz transition graph modulo 6 has spectral radius exactly 2, which drops to the golden ratio when the vertex for residue 4 is deleted.
Editorial extensions
If this is right
- The proportion of such restricted orbits decays exponentially at rate (φ/2)^m.
- Every positive cycle of T must visit residue class 2 mod 6.
- Residue class 2 mod 6 accounts for more than 18% of the steps in any positive cycle by flow conservation.
- Removing any single vertex from G' reduces the spectral radius strictly below 2.
- The spectral radii after single removals obey the strict hierarchy 1 < √2 < φ < 2.
Reading between the lines
- The explicit bijection Ψ_m from {1,…,6·2^m} to paths of length m may support exact counts under other forbidden-residue rules.
- Graph spectral methods of this form could be tested on modular restrictions for other iterative maps in number theory.
- The hierarchy of spectral radii after vertex deletions may suggest a natural ranking of residue importance in the dynamics.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper claims to prove that for each m≥1, exactly F(m+1) odd integers in {1,…,2^m} have Collatz orbits under the map T that avoid residue 4 mod 6 at steps 2 through m (with asymptotic proportion (φ/2)^m). The proof relies on the directed graph G of transitions modulo 6, its absorbing component G' on vertices {1,2,4,5} with spectral radius 2, and the subgraph obtained by deleting vertex 4 (on {1,2,5}) having spectral radius φ; the Fibonacci count is extracted from the resulting recurrence. An explicit bijection Ψ_m:{1,…,6·2^m}→P_m(G) (paths of length m) is constructed, and further claims are made that no vertex of G' is dispensable (with a strict hierarchy of spectral radii after single-vertex deletions) and that every positive cycle must visit residue 2 mod 6.
Significance. If the central count were correct, the result would give a precise combinatorial interpretation of a specific avoidance property in short Collatz trajectories via the spectral gap between the full component and the 4-avoiding subgraph, together with an explicit path bijection; the vertex-indispensability statements and the flow-conservation lower bound (>18%) on visits to 2 mod 6 in cycles would also be of independent interest for modular constraints on Collatz dynamics.
major comments (3)
- [Abstract] Abstract (main theorem statement): the claimed exact count of F(m+1) odd starters in {1,…,2^m} whose T-orbits avoid 4 mod 6 at steps 2..m is false for m=3. The four candidate odd integers are 1,3,5,7. Direct iteration yields: n=1 succeeds (T^2(1)≡1, T^3(1)≡2); n=3 fails (T^3(3)≡4); n=5 fails (T^2(5)≡4); n=7 succeeds (T^2(7)≡5, T^3(7)≡2). Thus only two successes, contradicting F(4)=3.
- [Proof of main theorem (via Ψ_m and spectral radius of G' minus vertex 4)] The bijection Ψ_m and the path-counting argument via the adjacency matrix of the subgraph on {1,2,5} (whose characteristic polynomial yields the Fibonacci recurrence) do not map onto the precise set of odd integers ≤2^m satisfying the avoidance condition for the window steps 2..m. The m=3 counterexample shows that the reduction from paths in G to the desired odd starters fails to produce the asserted cardinality.
- [Section deriving the count from the spectral gap] The extraction of the exact Fibonacci count from the spectral radius φ of the 4-avoiding subgraph assumes a direct correspondence between the number of length-(m-1) paths in that subgraph and the number of qualifying odd n≤2^m; the explicit enumeration for small m demonstrates that this correspondence does not hold for the stated avoidance window and parity restriction.
minor comments (1)
- [Abstract / Theorem statement] The precise definition of 'steps 2,…,m' (whether inclusive of the starting point or how the first iterate is indexed) should be stated explicitly in the theorem to avoid ambiguity in verification.
Simulated Author's Rebuttal
We thank the referee for the careful verification and the explicit m=3 counterexample. We have confirmed the calculations and agree that the main claim of exactly F(m+1) qualifying odd integers does not hold. The correspondence between paths in the 4-avoiding subgraph and the desired starters fails to produce the asserted exact count. We will revise the manuscript to remove or correct the incorrect theorem statement, the bijection claim, and the spectral-gap derivation while retaining any valid graph-theoretic observations.
read point-by-point responses
-
Referee: [Abstract] Abstract (main theorem statement): the claimed exact count of F(m+1) odd starters in {1,…,2^m} whose T-orbits avoid 4 mod 6 at steps 2..m is false for m=3. The four candidate odd integers are 1,3,5,7. Direct iteration yields: n=1 succeeds (T^2(1)≡1, T^3(1)≡2); n=3 fails (T^3(3)≡4); n=5 fails (T^2(5)≡4); n=7 succeeds (T^2(7)≡5, T^3(7)≡2). Thus only two successes, contradicting F(4)=3.
Authors: We acknowledge that the counterexample is correct and that the stated count is false for m=3. This reveals an error in the claimed exact enumeration. We will revise the abstract to remove the incorrect main theorem. revision: yes
-
Referee: [Proof of main theorem (via Ψ_m and spectral radius of G' minus vertex 4)] The bijection Ψ_m and the path-counting argument via the adjacency matrix of the subgraph on {1,2,5} (whose characteristic polynomial yields the Fibonacci recurrence) do not map onto the precise set of odd integers ≤2^m satisfying the avoidance condition for the window steps 2..m. The m=3 counterexample shows that the reduction from paths in G to the desired odd starters fails to produce the asserted cardinality.
Authors: The referee correctly identifies that Ψ_m counts paths in G but the reduction to the specific set of odd starters with the avoidance property for steps 2 to m is invalid. We will revise the proof section to excise the erroneous cardinality claim. revision: yes
-
Referee: [Section deriving the count from the spectral gap] The extraction of the exact Fibonacci count from the spectral radius φ of the 4-avoiding subgraph assumes a direct correspondence between the number of length-(m-1) paths in that subgraph and the number of qualifying odd n≤2^m; the explicit enumeration for small m demonstrates that this correspondence does not hold for the stated avoidance window and parity restriction.
Authors: We agree that no such direct correspondence exists, so the exact-count extraction is unjustified. The spectral radius provides an asymptotic rate but not the precise Fibonacci number for the claimed set. This section will be revised to reflect only the valid asymptotic statement. revision: yes
Circularity Check
No circularity: count derived directly from path enumeration in explicitly defined subgraph
full rationale
The derivation constructs the modular graph G on residues mod 6, identifies the absorbing component G' on {1,2,4,5}, removes vertex 4 to obtain the avoiding subgraph, and extracts the exact path count from its adjacency matrix (whose characteristic equation yields the Fibonacci recurrence). An explicit bijection Ψ_m is stated to map the relevant starters onto paths of length m. None of the six enumerated circularity patterns appear: the Fibonacci number is not presupposed in the definition of the graph or the avoidance condition, no parameter is fitted to data and then relabeled a prediction, and no self-citation supplies a uniqueness theorem or ansatz. The argument is self-contained against the graph adjacency and the bijection construction.
Assumptions & free parameters
assumptions (2)
- domain assumption The Collatz function T induces a well-defined directed graph of transitions on residue classes modulo 6.
- standard math Standard linear algebra results on the spectral radius of non-negative matrices (Perron-Frobenius theorem) apply to the adjacency matrix of the graph.
Cite this review
Pith. "Pith review of A Fibonacci theorem for Collatz trajectories via modular graph structure." pith.science (2026). https://pith.science/paper/S556K7H3
@misc{pith2026260602621,
author = {Pith},
title = {Pith review of: A Fibonacci theorem for Collatz trajectories via modular graph structure},
year = {2026},
howpublished = {\url{https://pith.science/paper/S556K7H3}},
note = {Machine review of arXiv:2606.02621}
}
abstract
Let $T(n)=n/2$ if $n$ is even and $T(n)=(3n+1)/2$ if $n$ is odd. We prove that for each $m\ge1$, exactly $F(m+1)$ odd integers in $\{1,\ldots,2^m\}$ have the property that their orbit under $T$ avoids the residue class $4\pmod6$ during steps $2,\ldots,m$, where $F(m+1)$ is the $(m+1)$-th Fibonacci number; the proportion decays at rate $(\varphi/2)^m$, $\varphi=(1+\sqrt{5})/2$. The proof uses the directed graph $G$ of Collatz transitions modulo $6$ and its unique absorbing strongly connected component $G'=G[\{1,2,4,5\}]$. Removing vertex $4$ from $G'$ yields a subgraph of spectral radius $\varphi$, against $\rho(G')=2$; the Fibonacci count follows from this spectral gap. We construct an explicit bijection $\Psi_m:\{1,\ldots,6\cdot2^m\}\to\mathcal{P}_m(G)$ onto the directed paths of length $m$ in $G$. We further show that no vertex of $G'$ is dispensable: removing any single vertex reduces the spectral radius strictly below $2$, with hierarchy $1<\sqrt{2}<\varphi<2$. In particular, every positive cycle of $T$ must visit residue class $2\pmod6$, and a flow conservation identity forces this class to account for more than $18\%$ of the steps in any such cycle.
Figures
Reference graph
Works this paper leans on
-
[1]
2015 , eprint =
Albert, Michael and Gudmundsson, Bjarki and Ulfarsson, Henning , title =. 2015 , eprint =
2015
-
[2]
and Lagarias, Jeffrey C
Bernstein, Daniel J. and Lagarias, Jeffrey C. , title =. Canad. J. Math. , volume =. 1996 , doi =
1996
-
[3]
On the existence of cycles of given length in integer sequences like
B. On the existence of cycles of given length in integer sequences like. Atti Accad. Naz. Lincei Rend. Cl. Sci. Fis. Mat. Nat. , volume =
-
[4]
, title =
Everett, Cornelius J. , title =. Adv. Math. , volume =. 1977 , doi =
1977
-
[5]
2026 , eprint =
Karras, Achilleas and de Weger, Benne , title =. 2026 , eprint =
2026
-
[6]
, title =
Lagarias, Jeffrey C. , title =. Amer. Math. Monthly , volume =. 1985 , doi =
1985
-
[7]
, title =
Lagarias, Jeffrey C. , title =. The Ultimate Challenge: The. 2010 , pages =
2010
-
[8]
, title =
Lagarias, Jeffrey C. , title =. 2006 , howpublished =
2006
Show all 11 references
-
[9]
Forum Math
Tao, Terence , title =. Forum Math. Pi , volume =. 2022 , doi =
2022
-
[10]
Acta Arith
Terras, Riho , title =. Acta Arith. , volume =. 1976 , doi =
1976
-
[11]
The Dynamical System Generated by the
Wirsching, G. The Dynamical System Generated by the. 1998 , doi =
1998
Reviewed June 29, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.