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

$G\subset \mathbf{C} , G= \{z\in \mathbf{C}| \overline{z}\in G\} = \overline{G}$, then for $f\in \mathcal{O}(G)$ it holds that:$$f(G\cap \mathbf{R}) \subset \mathbf{R} \Leftrightarrow \forall z\in G : f(\overline{z}) = \overline{f(z)}$$

This is an example in the script of our professor, however there is no proof for it and it hasn't been shown during the lectures either.

Proof : $"\Rightarrow " :$ Let $f(G\cap \mathbf{R}) \subset \mathbf{R}$, so all points on the real axis are mapped to the real axis. That means $f$ does not have any imaginary part, because then this would not be true anymore. Since f does not have any imaginary part. it follows also that : $f(\overline{z}) = \overline{f(z)}$

$"\Leftarrow" :$ Let $f(\overline{z}) = \overline{f(z)}$, so f can not have any imaginary part, otherwise this would not be true anymore. But if f doesn't have any imaginary part, then the map of the real axis is always on the real axis. So $f(G\cap \mathbf{R}) \subset \mathbf{R}$

What I mean with not having any imaginary part is that it is of the form $f(z) = z ; f(z)=2z; f(z) = 2z+z^{3}$ etc.

I am not sure if this is the right thought, and how to express this better.

Help is greatly appreciated.

share|cite|improve this question
Thanks for your attention. I tried to show it in my edited post. – PumaDAce Dec 6 '11 at 16:07
Yes, G is a domain. :) – PumaDAce Dec 6 '11 at 16:21
Ah, okay. I think it's good to mention that sort of thing, just for reassurance. – Dylan Moreland Dec 6 '11 at 16:25
I don't follow your $"\Rightarrow"$. If I write $f = u(x, y) + iv(x, y)$, then we find that $v(x, 0) = 0$ for all $x \in G \cap \mathbf{R}$. I don't see how to finish off from there -- maybe there are facts about harmonic functions that we could use? – Dylan Moreland Dec 6 '11 at 16:49
up vote 3 down vote accepted

($\Rightarrow$) I think it should be straightforward to check that the function $\overline{f(\bar z)}$ is holomorphic on $G$. Then $f(z) - \overline{f(\bar{z})}$ is a holomorphic function on $G$ which vanishes along an interval on the real axis.

($\Leftarrow$) I think you have the right idea, but we should write down some equations. Try to use the fact that $z = \bar z$ if and only if $z \in \mathbf R$; if you take $z \in \mathbf R \cap G$ and conjugate $f(z)$, what happens?

The problem is very related to the Schwarz reflection principle.

share|cite|improve this answer

:) :) :) Dylan Moreland :) :) :)

$$"\Rightarrow: "$$ If $f(G\cap\mathbf{R})\subset \mathbf{R}$, then for every $z_{0} \in \mathbf{R}$ we can write $f(z)= \sum_{k=0}^{\infty} a_{k}(z-z_{0})^{k}$, $a_{k}\in \mathbf{R}$. From this it follows immediately that $\overline{f(\overline{z})}=f(z)$ in a neighborhood of the real axis. Because of the identity theorem these two holomorphic functions must be the same everywhere.$$"\Leftarrow":$$ If $z\in \mathbf{R}$, $\overline{z}=z$. Then $f(\overline{z})=\overline{f(z)}= f(z)$ $\forall z\in \mathbf{R}$, so $f(z) \in \mathbf{R}$ , which means that $f(G\cap \mathbf{R})\subset \mathbf{R}$

Is this ok? Thank you for your help.

share|cite|improve this answer
No problem! I agree with your "$\Leftarrow$". For the other direction, it isn't clear to me how you get $a_k \in \mathbf R$. I think the holomorphicity should be obtainable without thinking about power series--let me try to work that out. The identity theorem is definitely what you want to use. – Dylan Moreland Dec 6 '11 at 19:06

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.