MathOverflow is a question and answer site for professional mathematicians. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Let $f \colon X \to Y$ be a flat morphism of schemes over $\mathbb{C}$. Suppose that $Y$ is normal and that the fibers over the closed points of $Y$ are all normal.

  1. Can I say something about the fibers over non-closed points?

  2. Is it true that $X$ is normal?

In the local setting, a positive answer to 2 can be found in Matsumura "Commutative ring theory", Theorem 23.9 and its corollary. However, the normality of all the fibers is required.

I am interested, in particular, in the case where the fibers over the closed point are all isomorphic to $\mathbb{P}^1_{\mathbb{C}}$.

share|cite|improve this question
Let's focus on the case where all the closed fibers are $\mathbf{P}^1$. I'm not sure if it's automatic, so I will also assume the generic fibre to be $\mathbf{P}^1$. Then, by flatness, all the fibres are (possible singular) curves over $\mathbf{C}$. The arithmetic genus is constant in the fibres. Also, the morphism is smooth over the set of closed points. Therefore it is smooth everywhere. In particular, all the fibres are $\mathbf{P}^1$. – Ariyan Javanpeykar Nov 30 '11 at 10:52
Thank you Ariyan. I don't understand why you say that all the fibers in this case are schemes over $\mathbb{C}$. – origal Nov 30 '11 at 11:29
Hi origal. I don't understand it either. Please ignore that. It should read, the fibre over $y$ in $Y$ is a (possibly singular) curve over the residue field $k(y)$. (For some reason, this morning, I thought this would always be $\mathbf{C}$.) The conclusion then reads that the fibre over $y$ is $\mathbf{P}^1_{k(y)}$. In view of Laurent Moret-Bailly's answer below, note that we do not use that $Y$ is normal. – Ariyan Javanpeykar Nov 30 '11 at 13:30
I think there is one more problem: what is constant in a flat family is the arithmetic genus, then, in this case $h^0(X_y,\mathcal{O}_{X_y})-h^1(X_y,\mathcal{O}_{X_y})=1$ (with $X_y$ fiber over $y$). But is it true that $h^0(X_y,\mathcal{O}_{X_y})=1$ if the variety $X_y$ is not defined over $\mathbb{C}$? I think we need $k(y)$ to be algebraically closed. – origal Nov 30 '11 at 15:17
up vote 8 down vote accepted

The fibers are normal. This follows (without the normality assumption on $Y$) from EGA IV (12.1.6) which says that the set those $x\in X$ where the fiber is normal is open, hence if its complement were nonempty, it would contain a closed point.

Hence, by your remark (or by EGA IV (6.8.3)), $X$ is normal.

share|cite|improve this answer
Perfect, thank you! – origal Nov 30 '11 at 11:36

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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