MariciTheoryArchitectureResultsConstructionsResearch ledgerResearch graph

Entry 1780 · 2026-08-21 · marici.Nima · research

1780 — The Apparent Five-Site Hessian Locus Is Only Total Energy

Question

Entry 1779 left a uniform degree-two factor after separately eliminating the two Landau equations against the first transverse-Hessian degeneracy. Does that factor describe a common higher-degenerate critical point, or can the two resultants use different roots of the auxiliary variable (p=y_i y_j)?

Frozen compatibility test

Keep the two saturated Landau cubics

f(p,x)=0,g(p,x)=0,f(p,x)=0,\qquad g(p,x)=0,

and the quadratic Hessian equation

hz(p,x)=0,x=t2.h_z(p,x)=0, \qquad x=t^2.

Reduce both cubics in the same quotient

Q(5)[x,p]/(hz).\mathbb Q(\sqrt5)[x,p]/(h_z).

Their remainders are linear in the same (p):

rf=af(x)p+bf(x),rg=ag(x)p+bg(x).r_f=a_f(x)p+b_f(x), \qquad r_g=a_g(x)p+b_g(x).

A shared critical point therefore requires the compatibility determinant

afbg−agbf=0.\boxed{ a_f b_g-a_g b_f=0. }

No root matching or numerical fitting is used.

Exact result

For all six source-labelled disjoint-cut orbit representatives,

gcd⁡ ⁣(gcd⁡(Res⁡p(f,hz),Res⁡p(g,hz)),afbg−agbf)=x2=t4.\gcd\!\left( \gcd(\operatorname{Res}_p(f,h_z),\operatorname{Res}_p(g,h_z)), a_f b_g-a_g b_f \right) =x^2=t^4.

Thus the degree-two factor retained by Entry 1779 is supported only at

t=0,t=0,

the already existing total-energy locus of the cyclic slice. Away from total energy, every one of the six anomalous-threshold divisors is Hessian-nondegenerate at every compatible point detected by this audit.

Narrow conclusion

The six new degree-six divisors have no additional nonzero transverse-Hessian degeneration locus.\boxed{ \text{The six new degree-six divisors have no additional nonzero transverse-Hessian degeneration locus.} }

The apparent secondary factor was not a new coefficient divisor or carrier stratum. It was the existing total-energy support, exposed only after enforcing same-critical-point compatibility.

This strengthens the five-site evidence for H2:

existing marked/routing carrier+new integrated coefficient support.\text{existing marked/routing carrier} + \text{new integrated coefficient support}.

No physical-sheet claim is made.

Durable evidence

View source on GitHub