Let $X$ be an affine smooth variety over the complex numbers, $X^{an}$ its associated smooth complex analytic space, and $\mathcal{O}$, resp. $\mathcal{O}^{an}$ the respective structure sheaves.

Is the natural map

$$\mathcal{O}(X)\to \mathcal{O}^{an}(X^{an})$$


Denoting by $f : X^{an}\to X_{Zar}$ the morphism of sites, this map is given as the composition

$$\Gamma(X_{Zar}, \mathcal{O})\to \Gamma(X^{an},f^{-1}\mathcal{O})\to \Gamma(X^{an},\mathcal{O}^{an})$$ and in this composition we know the last map is injective (see Serre’s GAGA paper, Prop. 10(b)).

What about the first map?

Morally, this map should send a regular function on $X$ to the holomorphic function it induces on $X^{an}$ in an evident way, so I would expect the answer to this question to be trivially yes, but I want to make sure I’m not missing anything.

If we call $\alpha$ the map $\Gamma(X_{Zar}, \mathcal{O})\to \Gamma(X^{an},f^{-1}\mathcal{O})$, then for any closed point $x\in X$, $\alpha(s)_x = \alpha_x(s_x)$, where $s\in \Gamma(X_{Zar}, \mathcal{O})$, $\alpha_x$ is the map induced on stalks, and $x$ is also regarded as a point of $X^{an}$, since analytification identifies closed points.

It is known that $\alpha_x$ is an isomorphism for all $x$ (see the discussion before Prop. 10 in loc cit), so this should be it. Am I right?

  • $\begingroup$ By the strong Nullstellensatz, the zero ideal in $\mathcal{O}_X(X)$ equals the intersection of all kernels of all evaluation $\mathbb{C}$-algebra homomorphism, $$\text{ev}_x:\mathcal{O}_X(X)\to \mathbb{C}.$$ Each of these homomorphisms factors through $\mathcal{O}_{X^\text{an}}(X^{\text{an}})$. $\endgroup$ – Jason Starr 3 hours ago

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Browse other questions tagged or ask your own question.