Let $\Omega\subset\mathbb{C}^n$ be a domain of holomorphy, $m<n$. Is $\Omega\cap\mathbb{C}^m$ necessarily a domain of holomorphy?

If true, it seems that it will be useful in inductive arguments. However, I could neither prove nor disprove this. It is true in some nice cases, e.g., if $\Omega$ is Reinhardt, but I have no idea in general.

Thanks in advance!


It need not be a domain to begin with, that is, the intersection could be disconnected. Furthermore, even if the original has smooth boundary, the intersection may not. But other than that, using the Hartogs-pseudoconvex definition (there exists a continuous plurisubharmonic exhaustion function), the intersection is Hartogs-pseudoconvex if the original was. (That a restriction of a plurisubharmonic function to a subspace is plurisubharmonic is obvious from the definition).

  • $\begingroup$ So every connected component of the intersection is a domain of holomorphy? $\endgroup$
    – Colescu
    Aug 22 '20 at 2:45
  • $\begingroup$ Yeah, but note that it could have weird behavior anyway. Many results in several complex variables are for domains with some sort of regularity on the boundary, and by intersection with a subspace, you are going to lose any regularity. You might get really weird boundaries in fact. $\endgroup$
    – Jiri Lebl
    Aug 23 '20 at 3:07

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.