3
$\begingroup$

Let $k$ be a field and let $X$ be a hyperbolic curve over $k$.

Then, there are only finitely many hyperbolic curves $Y$ over $k$ dominated by $X$.

I know this statement holds over $k=\mathbf{C}$. In particular, it holds over $k=\overline{\mathbf{Q}}$.

Does it hold over any field $k$?

What about a number field?

$\endgroup$
  • $\begingroup$ What is a hyperbolic curve over $k$ ? $\endgroup$ – Georges Elencwajg Oct 22 '12 at 18:42
  • $\begingroup$ A smooth projective geometrically connected curve of genus $\geq 2$. $\endgroup$ – Harry Oct 22 '12 at 19:05
  • $\begingroup$ Thanks. Where did you see that definition? $\endgroup$ – Georges Elencwajg Oct 22 '12 at 19:22
  • $\begingroup$ I can't remember...I think sometimes people also use it for an integral curve whose normalization is a smooth projective geometrically connected curve of genus $\geq 2$. $\endgroup$ – Harry Oct 22 '12 at 19:32
2
$\begingroup$

The answer is yes if your curves are projective, see E. Kani: Bounds on the number of nonrational subfields of a function field. Invent. Math. 85 (1986), no. 1, 185–198.

Over an algebraically closed field, a hyperbolic curve is either a non-empty open subset of a projective curve of genus $g\ge 2$, or an elliptic curve minus at least one point, or the projective line minus at least three points. Equivalently, this is a smooth geometrically connected curve (not necessarily projective) with finite automomorphism group.

$\endgroup$
  • $\begingroup$ A very important word here is separable. In characteristic $p > 0$, you can pre- or postcompose with arbitrary powers of Frobenius to get infinitely many morphisms. The cited article only covers separable maps. $\endgroup$ – Remy Nov 6 '17 at 7:01

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.