Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. 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

After to post this question: Help with a proof in Hartshorne's book I realized that I have another doubt in this proof:

What I know is

$\varphi_p(s_p)=t_p\implies (\varphi_{V_p}(s(P))_p=t_p$

I don't understand why $(\varphi_{V_p}(s(P))_p=(t_{|V_P})_p$


Instead of $\varphi(V_P)$, to simplify notation I write $\varphi_{V_P}$

Sorry to post the same proof again

Thanks a lot.

share|cite|improve this question
Recall the definition of the stalk - it answers your question directly. My advice is not to read this Proposition step my step as if it is was a purely algebraic argument. Instead, do it on your own and use pictures. Really, you can prove this Proposition just using pictures, and thereby gain a lot of intuition for the subject. I would explain it to you if we were in front of a blackboard. :) – Martin Brandenburg Sep 18 '13 at 12:14
@MartinBrandenburg Unfortunately I don't have maturity in this field yet to use pictures. – user42912 Sep 19 '13 at 8:08
@MartinBrandenburg maybe there is some typo or simplification in the notation I don't see? – user42912 Sep 19 '13 at 8:09
@MartinBrandenburg please can you see my answer? – user42912 Sep 19 '13 at 12:08

Consider the morphisms

  • $\rho_{X/V_P}:\mathcal{G}(X)\to\mathcal{G}(V_P)$
  • $\sigma_{V_P/P}:\mathcal{G}(V_P) \to \mathcal{G}_P$
  • $\sigma_{X/P}:\mathcal{G}(X)\to\mathcal{G}_P$

It follows from the definition of a stalk that $\sigma_{X/P}=\sigma_{V_P/P}\circ\rho_{X/V_P}$ because $V_P$ and $X$ are both open neighborhoods of $P$. Look up the definition of a direct limit to confirm this, if you like.

In other words, the above statement means $t_P=(t|_{V_P})_P$. Now given what you know, this yields what you want to understand.

Edit I would write it down as follows: We may replace $V_P$ by $V_P\cap U$, then $V_P$ is an open subset contained in both $V_P$ and $U$. Hence, we have $t_P=(t|_{V_P})_P$. Thus, $\varphi_P(s_P)=t_P$ implies $(\varphi_{V_P}(s(P)))_P=t_P=(t|_{V_P})_P$.

share|cite|improve this answer
Thank you for your help, please can you see my answer? – user42912 Sep 19 '13 at 12:08

Lema 1

Let $V\subset U$ and $t\in \mathcal F(U)$, we have $(t_{|V})_p=t_p$


$(t_{|V})_p=t_p \Leftrightarrow \exists \ W\subset V\cap U=V;\ t_{|W}=(t_{|V})_{|W}$, but


Proof of the question:

$\varphi_P(s(P))=t_P\implies (\text{by definition of}\ \varphi_P)\\ (\varphi_{V_P}(s(P)))_P=t_P\implies\\ \exists\ W\subset V_P\cap U=V_P; (\varphi_{V_P}(s(P)))_{|W}=t_{|W}\implies \ (\text{since} \ t_{|W}=(t_{|V_P})_{|W})\\(\varphi_{V_P}(s(P)))_{|W}=(t_{|V_P})_{|W}\implies\\ ((\varphi_{V_P}(s(P)))_{|W})_P=((t_{|V_P})_{|W})_P\implies\text{(Lemma 1)}\\ (\varphi_{V_P}(s(P)))_P=((t_{|V_P}))_P$

share|cite|improve this answer
This looks correct, but it does use quite a large amount of brackets. I think this can be seen very easily in diagram language, while the above is hard to read. – Jesko Hüttenhain Sep 20 '13 at 6:28

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.