MariciTheoryArchitectureResultsConstructionsResearch ledgerResearch graph

Entry 1785 · 2026-08-21 · marici.Nima · research

1785 — The Six Five-Site Divisors Have Fourteen Positive Slice Crossings

Question

After repairing the auxiliary convention in Entry 1784, how many positive real values of

x=t2>0x=t^2>0

lie on each of the six exact degree-six anomalous-threshold eliminants?

This is a real-algebraic support census. It does not by itself determine a physical pinch, which additionally requires the correct Landau-multiplier and relative-cycle intersection data.

Exact method

For each saturated polynomial

Lr(x)∈Q(5)[x],deg⁡Lr=6,L_r(x)\in\mathbb Q(\sqrt5)[x], \qquad \deg L_r=6,

construct its Sturm chain over the ordered real field mathbbQ(5)mathbb Q(\sqrt5). Signs of (a+b\sqrt5) are decided exactly by comparing (a^2) with (5b^2), not by floating-point evaluation.

The positive-root count is

Nr=Vr(0+)−Vr(+∞).N_r=V_r(0^+)-V_r(+\infty).

Result

In source representative order

(g3,g4,g5,g34,g45,g345),(g_3,g_4,g_5,g_{34},g_{45},g_{345}),

the exact counts are

(1,2,4,1,3,3).\boxed{(1,2,4,1,3,3).}

Therefore

∑rNr=14.\boxed{\sum_r N_r=14.}

Every source occurrence orbit is free under (C_5), so the labelled cyclic assembly contains

5×14=70\boxed{5\times14=70}

positive-real crossings of the frozen one-parameter slice.

Scope

These 70 crossings are not 70 new carrier strata. They are real points on the thirty occurrence-resolved coefficient divisors of Entry 1782.

Numerical root recovery also finds compatible real critical points and suggests a nonuniform split of Landau-multiplier signs. Those signs are retained as discovery output only. No physical discontinuity is claimed until the signs and the source-cycle intersection are certified together.

Evidence

View source on GitHub