On the canonical degrees of curves in varieties of general type (Q1930895): Difference between revisions

Conjecturally, surfaces of general type contain only finitely many rational or elliptic curves. More generally, for every \(g\) the curves of genus at most \(g\) on a variety of general type \(X\) should form a bounded family. There is also an effective version of this conjecture stating the following: Given a smooth projective variety \(X\) of general type there are constants \(A,B\) and a closed algebraic subset \(Z\subset X\) such that for every finite morphism \(f\) from a curve \(C\) to \(X\) with \(f(C)\not\subset Z\) we have \(\deg f^*K_X\leq A(2g_C-2)+B[C:f(C)]\). This conjecture is proved for compact locally symmetric spaces with \(A=\dim X\), \(B=0\) and \(Z=\{\;\}\). Using certain Shimura varieties it is also shown that the conjecture can not hold in general with \(A<\dim(X)\).