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

Let $C$ be a Grothendieck-site with finite limits (perhaps I have to assume that the topology it defines is subcanonical, but it would be nice if I wouldn't have to) and let $T=Shv(C)$ denote the category of sheaves on it.

Let $f:F\to G$ be a morphism in $T$ and suppose that for all $X\in T$ there exists objects $Y_{i,X}\in T$ and an epimorphism $\coprod Y_{i,X}\to X$ in $T$ such that the map of sets \begin{equation} (*)\hspace{3em}Hom_T(Y_{i,X},f):Hom_T(Y_{i,X},F)\to Hom_T(Y_{i,X},G) \end{equation} is surjective. How can I deduce that $f:F\to G$ is an epimorphism of sheaves?

Remark: I am pretty sure, that this is true since if for every representable $X$ the $Y_{i,X}$ would be representable, they would cover $X$ and condition (*) would imply that $f$ is an epimorphism of sheaves. However, I cannot show this when the $Y_{i,X}$ are not representable. I have tried to write them as a colimit of representables $Y_{i,X}=colim_\alpha Y_{i,X}^\alpha$ and use $Hom_T(Y_{i,X},f)\cong \lim_\alpha Hom_T(Y^\alpha_{i,X},f)$ but I don't get further.

share|cite|improve this question
up vote 2 down vote accepted

Well, if the condition holds for all $X$, then in particular it holds for $X = G$. So choose $q_i : Y_i \to G$ such that $q : \coprod_i Y_i \to G$ is an epimorphism and $(*)$ holds. Then there must exist $p_i : Y_i \to F$ such that $q_i = f \circ p_i$, by hypothesis, and so if $p : \coprod_i Y_i \to F$ is the induced morphism, then $q = f \circ p$. But $q$ is an epimorphism, so $f$ must be as well.

Note that the above argument does not depend on the fact that we are working in a topos. Rather, it is the converse that requires something extra!

share|cite|improve this answer
Ah, yes, thanks, I was a little blind. Unfortunately, I can't up vote your answer. – sopot Dec 2 '13 at 20:35

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.