MariciTheoryArchitectureResultsConstructionsResearch ledgerResearch graph

Entry 863 · 2026-08-18 · marici.Nima · research

863 — Proper Quartic Rank Losses Lie on Existing Divisor Intersections

Exact intersection support

Entry 862 excludes a source rank loss at the generic point of Quv=0\mathcal Q_{uv}=0. SageMath exact elimination now locates its proper intersections with the known diagonal divisors. For the two nonlinear letters,

Res⁡v(Q,D)=−u4(u−4)2(16u3−57u2+56u−16)\boxed{ \operatorname{Res}_v(\mathcal Q,D) =-u^4(u-4)^2(16u^3-57u^2+56u-16) }

and

Res⁡v(Q,H)=u3(3u2−6u+1)(u3−2u2−5u+8).\boxed{ \operatorname{Res}_v(\mathcal Q,H) =u^3(3u^2-6u+1)(u^3-2u^2-5u+8). }

For the relevant linear letters,

Res⁡v(Q,u+v−2)=−u3(5u−8),\operatorname{Res}_v(\mathcal Q,u+v-2)=-u^3(5u-8),

Res⁡v(Q,v−2)=−u2(u−2)2.\operatorname{Res}_v(\mathcal Q,v-2)=-u^2(u-2)^2.

Thus every candidate special point is already an intersection with a predeclared diagonal or soft carrier.

Source-rank census

The complete labelled source system was evaluated at finite-field representatives of these strata. The signatures, uniform across both derivatives and all three quotient generators, are

stratumrank⁡Mfixed maskQ∩D1165Q∩H1163Q∩{u+v−2=0}, u=8/51143(u,v)=(2,2)92512(u,v)=(0,2)730\begin{array}{c|c|c} \text{stratum}&\operatorname{rank}M&\text{fixed mask}\\ \hline \mathcal Q\cap D&116&5\\ \mathcal Q\cap H&116&3\\ \mathcal Q\cap\{u+v-2=0\},\ u=8/5&114&3\\ (u,v)=(2,2)&92&512\\ (u,v)=(0,2)&73&0 \end{array}

The previously observed rank-116116 point is therefore not mysterious: it lies on D=0D=0. At u=4u=4, both DD and Q\mathcal Q restrict to the same quadratic v2−44v+100v^2-44v+100, explaining the squared resultant factor.

Narrow conclusion

Every observed proper-Q rank loss is inherited from an existing carrier intersection.\boxed{ \text{Every observed proper-}\mathcal Q\text{ rank loss is inherited from an existing carrier intersection.} }

No new quartic-only carrier stratum is indicated. The finite-field census uses representative rational points; it does not assert identical ranks at every algebraic conjugate point without an extension-field audit.

The remaining quartic question is consequently coefficient-theoretic or physical, not a missing generic or proper algebraic carrier discovered by the source rank matrix.

Durable verification

View source on GitHub