# Characterisation of coherent sheaves on an algebraic variety

The Wikipedia article on coherent sheaves makes the following claim (without any reference), which I had trouble proving or finding a reference for: on an algebraic variety X (or I guess possibly even on a locally noetherian scheme), the coherent sheaves can be defined as the smallest class of sheaves of $\mathcal{O}_X$-modules with the following two properties:

i) the sheaf $\mathcal{O}_X$ is itself coherent;

ii) if, in a short exact sequence of sheaves, two of the sheaves are coherent, then so is the third.

I'm skeptical, but I would still like to know if this is true. If so, does anyone know a reference?

-
Wouldn't the first Chern class of every sheaf in that class be zero, since $c_1$ is additive over short exact sequences? –  MartinG May 20 '10 at 14:25
Oh wow, I feel silly. You should post that as an answer so I can accept it. –  Dan Petersen May 20 '10 at 14:43
It isn't silly, after all it was claimed at Wikipedia, and you were skeptical.. –  MartinG May 20 '10 at 18:21
I think if you replace (ii) with the requirement that the kernel and cokernel of any morphism between coherent sheaves is coherent, then it works. –  Sándor Kovács Jun 19 '11 at 2:34

This is false. Every sheaf in that class would have zero first Chern class, since $c_1$ is additive over short exact sequences.
Perhaps the original Wikipedia article meant to have "$\mathcal{O}_{Z}$ is coherent for every subvariety $Z\subseteq X$" in place of (i). Then the fact that the smallest class of sheaves satisfying (i) and (ii) is the category of coherent sheaves is an application of the usual d&eacute;vissage-type argument. –  Mike Roth May 26 '10 at 3:20