Table of Contents
Fetching ...

Rational lines on cubic hypersurfaces II

Julia Brandes, Rainer Dietmann, David B. Leep

TL;DR

This work sharpens the bound for the existence of $K$-rational lines on cubic hypersurfaces over number fields by developing an adaptive non-real quadratic-extension strategy. A central technical advance shows that for a diagonal quadratic form in $n=2k+1$ variables there exists a non-real extension $L/K$ making the form vanish on an affine $L$-linear space of dimension at least $k$, enabling a reduction to a smaller-dimensional search for a line. Applying this to cubics yields: if $n\ge 35$ in general, $n\ge 33$ for imaginary quadratic $K$, or $n\ge 31$ when $K=\mathbb{Q}$, then $C=0$ contains a $K$-rational line, improving Wooley's bound; the paper also proves sharpness of the dimension bound and discusses a Leep-based variant for imaginary quadratic fields. The results extend Birch–Wooley type rational-space conclusions to broader settings and demonstrate a robust local-to-global framework for quadratic forms under quadratic extensions.

Abstract

We show that any rational cubic hypersurface of dimension at least 33 defined over a number field $K$ vanishes on a $K$-rational projective line, reducing the previous lower bound of Wooley by two. For $K=\mathbb Q$ we can reduce the bound to 29. The main ingredients are a result on linear spaces on quadratic forms over suitable non-real quadratic field extensions, and recent work of Bernert and Hochfilzer on cubic forms over imaginary quadratic number fields for the rational case.

Rational lines on cubic hypersurfaces II

TL;DR

This work sharpens the bound for the existence of -rational lines on cubic hypersurfaces over number fields by developing an adaptive non-real quadratic-extension strategy. A central technical advance shows that for a diagonal quadratic form in variables there exists a non-real extension making the form vanish on an affine -linear space of dimension at least , enabling a reduction to a smaller-dimensional search for a line. Applying this to cubics yields: if in general, for imaginary quadratic , or when , then contains a -rational line, improving Wooley's bound; the paper also proves sharpness of the dimension bound and discusses a Leep-based variant for imaginary quadratic fields. The results extend Birch–Wooley type rational-space conclusions to broader settings and demonstrate a robust local-to-global framework for quadratic forms under quadratic extensions.

Abstract

We show that any rational cubic hypersurface of dimension at least 33 defined over a number field vanishes on a -rational projective line, reducing the previous lower bound of Wooley by two. For we can reduce the bound to 29. The main ingredients are a result on linear spaces on quadratic forms over suitable non-real quadratic field extensions, and recent work of Bernert and Hochfilzer on cubic forms over imaginary quadratic number fields for the rational case.
Paper Structure (5 sections, 5 theorems, 18 equations)

This paper contains 5 sections, 5 theorems, 18 equations.

Key Result

Theorem 1

Let $K \mid \mathbb Q$ be a number field, let $n \in \mathbb N$ and let $C \in K[X_1, \ldots, X_n]$ be a cubic form. If $n \ge 2\delta_K+3$, then $C$ vanishes on a $K$-rational projective line.

Theorems & Definitions (7)

  • Theorem 1
  • Theorem 2
  • Theorem 3
  • Lemma 2.1
  • proof
  • Lemma 2.2
  • proof