the power-of-three conjecture

formerly and briefly styled “the Fong Conjecture” — a misattribution, corrected: Danielle Fong never stated it; the statement was machine-proposed. Epitaph in the dustbin.

status: refuted · posed 2026-07-20 ≈11:00 · refuted 2026-07-20 ≈14:00

The degree of every étale polynomial endomorphism of ℂ³ is a power of 3.

Refuted by an explicit degree-4 map (“F₄”) from Alexis Gallagher's weighted-lift construction (jacobianfun.org), which produces a counterexample for every generic fiber degree n ≥ 3 — an atlas of 98 of them, degrees 3 through 100, is published there. We transcribed F₄ and verified it on this site's own tools: det JF₄ ≡ 1 exactly; generic fiber degree 4; four distinct preimages over a rational point. The conjecture named its own attack surface (“a quartic fold”) when posed; the attack landed the same afternoon. This page promised to record the result either way. Here it is.

what it became

theorem (same day)

The degree spectrum of étale polynomial endomorphisms of ℂ³ is exactly {1} ∪ {3, 4, 5, …}.

Upper half (every n ≥ 3 occurs): Gallagher's weighted-lift family — a one-variable seed p with p(0)=0, p(1)=−c, ∫₀¹p = 0 yields fiber degree deg p + 1. Lower half (n = 2 never occurs): a two-sheeted cover carries a mirror symmetry; symmetric covers are Galois; Galois + constant Jacobian forces an automorphism (Campbell 1973) — recorded on this project hours before the refutation arrived. The conjecture was wrong; the question was right; the two halves of the answer came from different people working the same 24 hours. That is the system functioning.

Ledger and machine verification: jacobianconjectures.com (check C9 = F₄) · constitution: fablehaven.cc — corrections are recorded, never erased.