scientific article; zbMATH DE number 1341991

From MaRDI portal
Publication:4263558





This paper considers three questions on curves in projective 3-space over an algebraically closed field \(k\). For clarity the questions are formulated as conjectures. The paper makes observations about the conjectures, gives examples, and identifies the ``smallest particular cases which remain open. The conjectures are: Conjecture A. Assume \(\operatorname {char} k=0\). Then the smooth specialization of a complete intersection curve is a complete intersection. This is false in characteristic \(p\) [see \textit{N. Mohan Kumar}, Invent. Math. 104, No.~2, 313-319 (1991; Zbl 0739.14013], but Kumar's methods do not extent to characteristic 0. In the present paper it is shown that, in characteristic 0, any smooth specialization of a complete intersection of type \((a,b)\) with \(a \leq 4\) is a complete intersection, except possibly for the case \((4,5)\), which remains open. Conjecture B. Assume \(\operatorname {char} k= 0\). If a family of curves \(C_t\) lying on normal surfaces \(X_t\) of degree \(s=s(C_t)\) specializes to a curve \(C_0\) with \(s(C_0)<s\), then the degree of the curves is bounded by \(s^2\). It is shown that conjecture B implies conjecture A, and examples are given as evidence for B. Conjecture C. There exist families of curves in characteristic \(p\) which do not lift to characteristic 0. It is shown that this is implied by a conjecture B\(^\prime\), closely related to B, and evidence for B\(^ \prime\) is given. For the entire collection see [Zbl 0913.00044].











This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4263558)