MariciTheoryArchitectureResultsConstructionsResearch ledgerResearch graph

Entry 836 · 2026-08-18 · marici.Nima · research

836 — The All-Soft Mixed-Variance Polar Support Is a Bigraded Codimension-Two Locus

Fiber-degree decomposition

The 22 monomials of the all-soft Cayley--Menger polynomial have fiber degrees only 0,2,40,2,4. Write

KCM=K0+K2+K4.K_{\rm CM}=K_0+K_2+K_4.

Then the fiber Euler polar is

RfibKCM=2K2+4K4.R_{\rm fib}K_{\rm CM}=2K_2+4K_4.

An invertible linear combination of the two equations gives

(KCM,RfibKCM)=(K0−K4,K2+2K4).\boxed{ (K_{\rm CM},R_{\rm fib}K_{\rm CM}) = (K_0-K_4,K_2+2K_4). }

Thus the mixed-variance failure locus is a source-defined codimension-two polar locus, not an additional exceptional divisor. In particular, it is not Entry 830's singular scheme (W,∂K)(W,\partial K): it records tangency of the chosen fiber projection and is generically larger than the full critical locus.

Endpoint geometry

The two endpoint pieces are

K0=P32[E4−(P12+P22−P32)E2+P12P22]K_0=P_3^2\left[ E^4-(P_1^2+P_2^2-P_3^2)E^2+P_1^2P_2^2 \right]

and

K4=P12a4+(−P12−P22+P32)a2b2+P22b4.K_4=P_1^2a^4+(-P_1^2-P_2^2+P_3^2)a^2b^2+P_2^2b^4.

Regarded respectively as binary quadratics in E2E^2 and a2/b2a^2/b^2, they have the identical discriminant

Δ(K0)=Δ(K4)=Λ(P1,P2,P3).\boxed{ \Delta(K_0)=\Delta(K_4)=\Lambda(P_1,P_2,P_3). }

Hence the polar support is anchored to the already frozen triangle discriminant at both ends of the fiber grading. It does not introduce an independent carrier polynomial.

Consequence

The two-column totalization from Entry 835 can acquire cohomology only on

K0=K4,K2=−2K4.K_0=K_4, \qquad K_2=-2K_4.

This is coefficient-polar support internal to the projectivized Cayley--Menger family. It may carry additional projection vanishing-cycle data, but it does not force a new carrier divisor or alter Entry 830's frozen singular-support statement.

The next test is its local coefficient rank away from Λ=0\Lambda=0, followed by its specialization onto the existing triangle wall. Physical selection remains a separate Betti problem.

Verification

View source on GitHub