On set-theoretic intersections (Q5902890): Difference between revisions

In his book ''Ample subvarieties of algebraic varieties'' [Lect. Notes Math. 156 (1970; Zbl 0208.489)], p. 126, \textit{R. Hartshorne} posed the following problem: Generalize Kneser's result to higher dimension. Is it true, for example, that any connected subvariety \(Y\subset {\mathbb{P}}^ n_ k\) of dimension s is set-theoretic intersection of at most 2n-2s-1 hypersurfaces? Partial answers were given by P. Schenzel, W. Vogel, T. Schmitt and P. Newstead. - In the paper under review, the author improves the previous results, by showing, that: ''For each n, t, with \(n>2t\), there exist a connected subvariety \(V_{n,t}\subset {\mathbb{P}}^ n_ k\) (k algebraically closed of arbitrary characteristic) of pure dimension t, which is set-theoretic intersection of not fewer than \(n+1-[n/t]\) hypersurfaces (where [n/t] denotes the largest integer \(\leq n/t)\).