I'm trying to understand the claim that the moduli space of stable rank 2 vector bundles on a (general?) cubic 3-fold, say $X$, with $deg c_2=6$ and $det=O_X(2)$ is of dimension 9.

I belive what I can do, is to compute the tangent space at a general point, says $E$, and try to compute its dimension, and I seem to recall that this can be done by computing the $H^1$ of a certain bundle ($End(E)?$) but I haven't managed to find any specific reference for that fact.

Is it correct? And if so, where does that come from?


1 Answer 1


The Zariski tangent space of the moduli space of stable sheaves at a point $[F]$ for a stable sheaf $F$ can be canonically identified with $Ext^1(F,F)$. Now if $F$ is locally free, then this space is just $H^1(X,\mathcal{E}nd(F))$.

You can read about all that in Huybrechts-Lehn: The Geometry of Moduli Spaces of Sheaves.

Where did you read that the moduli space you investigate is 9-dimensional?

  • $\begingroup$ Thank you! I read it here math.polytechnique.fr/~voisin/Articlesweb/fibresaj.pdf on page 11. $\endgroup$
    – PTV
    May 20, 2015 at 16:03
  • $\begingroup$ Do you see an easy way prove that such a vector bundle $E$ in the moduli space satisfies $H^1(X, End(E))$ is 9-dimensionnal. I gave this some thought, and tried to use Riemann-Roch, but i don't have the right result. $\endgroup$
    – PTV
    May 22, 2015 at 14:15
  • $\begingroup$ I thought about that myself. I mean there is a fairly explicit description of the bundle, but i couldn't do it either. Maybe one should try and write this bundle as an extension using the Serre correspondence. Maybe you should ask this in a new question. $\endgroup$
    – Bernie
    May 22, 2015 at 16:50

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .